institute for systems, informatics and...

joint research centreEUROPEAN COMMISSION

2 0 0 0

ANNUAL REPORTInstitute for Systems, Informatics and Safety

Report EUR 19743 EN

EUROPEAN COMMISSIONJOINT RESEARCH CENTRE

European Commission

Joint Research Centre (DG JRC)

Institute for Systems, Informatics and Safety (ISIS)

Editor:

A.M. Morrissey

Graphic design and layout:

R. Sánchez • Information and Public Relations Unit • JRC Ispra

Legal Notice

Neither the European Commission nor any person

acting on behalf of the Commission is responsible for the use

that might be made of the information contained in this publication.

EUR Report 19743 EN

Luxembourg: Office for Official Publications

of the European Communities, 2001

ISBN 92-894-0602-X

© European Communities, 2001

Printed in Italy

Institute for Systems,Informatics and Safety

Institute for ReferenceMaterials and Measurements

Institute forTransuraniumElements

Institute forProspectiveTechnological Studies

InstituteforAdvancedMaterials

The Institute for Systems, Informatics and Safety (ISIS) is one of eight institutes that constitute the EuropeanCommission’s Directorate-General Joint Research Centre (DG-JRC). Situated in Ispra, Northern Italy, ISIS supportsEU policies with systems-oriented research in areas where the safety and security of the citizen are of concern.In particular, ISIS focuses on the following policy support themes: nuclear safeguards, non-proliferation andinternational humanitarian security; cyber-security, knowledge access and assessment, the fight against fraud;and natural and technological risks and emergencies.

Institute for Systems,Informatics and Safety

Environment InstituteSpaceApplicationsInstitute Institute for Health

and Consumer Protection

How to contact ISIS

Dr. Anne-Marie Morrissey

“Communications and Strategy”

European Commission DG-Joint Research Centre

Institute for Systems, Informatics and Safety (ISIS)

T.P. 361

I-21020 Ispra (Varese) • Italy

Tel: +39 0332 789322/9511

Fax: +39 0332 785238

E-mail: [email protected]

http://www.jrc.cec.eu.int/isis/

CONTENTS

Foreword

Organisation of ISIS

Operating mode

Policy support themes

Nuclear Safeguards, Non-proliferation and

International Humanitarian Security

1. Safeguarding Against the Proliferation of Nuclear Weapons

2. Humanitarian Demining

Cyber-security, Knowledge Access and Assessment,

the Fight against Fraud

3. Dependability of Information Technology Systems

4. Networks, Multimedia and Education

5. Medical Imaging - Data Systems

6. Applied Statistics

7. Knowledge Assessment Methodologies

8. Fight against Fraud

Natural and Technological Risks and Emergencies

9. Safety and Emergency Management of Man-made and Natural Hazards

10. Decision Support for Risk and Emergency Planning

11. Aviation Safety

12. Seismic Protection for Civil and Cultural Heritage

13. Safety in Transient Structural Dynamics

14. Nuclear Safety

Resources

Staff

Budget

ISIS Contact Points

4

5

6

8

9

10

24

27

28

30

38

40

44

46

54

56

66

72

78

88

94

108

108

108

109

Contents 3

4 Foreword

The year 2000 offered the Institute for Systems, Informatics and Safety (ISIS) not

only the opportunity to refocus and concentrate its activities, but the possibility to

strongly re-affirm its role in serving European Union policy makers.

In January 2000, the Communication Towards a European Research Area1 (initiative

of the European Commissioner for Research Mr. Philippe Busquin) outlined the JRC’s

contribution to the European Research Area in terms of supporting the development

of a scientific and technical reference area for policy support. In June 2000, A review

of the Joint Research Centre by an independent High Level Panel, chaired by Viscount

E. Davignon, outlined several priorities for non-nuclear research at the JRC: (1) security for the European

citizen, including health, food, environment and privacy; (2) security for economic issues including new

crimes related to the information society and intellectual and industrial property rights; (3) enlargement;

and (4) the common foreign and security policy. With this in mind, the mission statement of ISIS focuses

on supporting EU policies with systems-oriented research in areas where the safety and security of

the citizen are of concern in the following policy support themes:

• nuclear safeguards, non-proliferation and international humanitarian security;

• cyber-security, knowledge access and assessment, the fight against fraud;

• natural and technological risks and emergencies.

To strengthen the Institute’s fulfilment of its new mission statement and to provide a better clustering

of expertise, the humanitarian demining technologies work was transferred from the Space Applications

Institute to ISIS in October 2000.

ISIS’s role is to provide a service to meet the requirements of its customers – other Directorates-

General (DGs) and services of the European Commission – thereby providing a direct contribution to

European policy processes. This service covers the full cycle of policy making from the conception of

a new policy through support in its development and implementation to its monitoring and evaluation.

Consistent with Commissioner Busquin’s project on the European Research Area, added benefit and

value derives from ISIS being able to call upon the expertise of many research establishments,

throughout the European Union, due to its extensive involvement in networks and collaborative

actions, and from its independence of national and private interests as a European research organi-

sation. In such a manner, ISIS brings a greater critical mass to bear on its support to policy making

and provides a gateway to the world of science and technology for its DG customers.

David R. WilkinsonDirector

Institute for Systems, Informatics and Safety

FOREWORD

MISSION STATEMENT OF ISIS

ISIS supports EU policies with systems-oriented research in areas where the safety and security of the citizen

are of concern. The application areas include:

• nuclear safeguards, non-proliferation and international humanitarian security;

• cyber-security, knowledge access and assessment, the fight against fraud;

• natural and technological risks and emergencies.

Its prime objectives are to develop techniques for the assessment of risk in complex systems and to apply in-

formation, communication and engineering technologies for improving their reliability, safety and security.

The strong cross-fertilisation between its nuclear and non-nuclear activities strengthens the expertise it can

bring to the benefit of customers in both domains.

1 “Towards a European Research Area”, COM(2000)6, January 2000.

ORGANISATION OF ISIS

The diagram reflects the organisational structure ofISIS and the main focus of the activities of each unit.Some overlap between the competencies may occur.

Organisation of ISIS 5

Nuclear safeguards, non-proliferation and international humanitarian security

Cyber-security, knowledge access andassessment, the fight against fraud

Natural and technological risks and emergencies

Quality andManagement Support

...............

Deputy DirectorM. Cuypers

Safety in StructuralMechanicsM. Géradin

Safeguards andVerification Techniques

M. Cuypers

Institute DirectorD.R. Wilkinson

Reliable InformationTechnologies

F. Sorel

Methodologies forInformation Analysis

M. Dowell

Risk Management andDecision Support

A.C. Lucia

Technologies forDetection and Positioning

A.J. Sieber

Management support team

Unit Heads: Units:

1. Marc Cuypers (Deputy Director) Safeguards and Verification Techniques2. Alois J. Sieber Technologies and Detection Positioning3. Fernand L. Sorel Reliable Information Technologies4. Martyn Dowell Methodologies for Information Analysis5. Alfredo C. Lucia Risk Management and Decision Support6. Michel Géradin Safety in Structural Mechanics7. Horst Weisshäupl Nuclear Safety

1. 2. 3.

4. 5. 6. 7.

Operating mode

OPERATING MODE

6

The Communication Towards a European Research Area

(initiative of the European Commissioner for ResearchMr. Philippe Busquin) embraces, in particular, the follow-ing: networking, improved use of infrastructures, trainingand mobility and the establishment of a common scien-tific and technical reference system for the implemen-tation of policies – all of which involve partnerships,collaborations, the pooling of complementary expertise,information exchange, mutual learning and knowledgesharing. ISIS is not only active in promoting networks,operating several successful research infrastructures,providing research training for researchers and playinga role in developing a common scientific reference sys-tem, but it is also very dynamic in encouraging collabo-rations and partnerships and in creating a wider cultureof research and innovation by earmarking a portion ofits budget for exploratory research.

NETWORKING

ISIS currently utilises networking as the preferred for-mula when undertaking ambitious research activities.Up until now, ISIS’s networking has been mainly throughcompetitive projects won in consortiums through theindirect actions of the Framework Programme. Manynetworks have been centred around infrastructuressuch as ELSA (European Laboratory for Structural Ass-essment). While ISIS continues to maintain a strongcommitment to competitive work and facility-orientednetworks, new mechanisms of networking will be ex-plored in order to bring the expertise of the Europeanscientific community to bear on specific problems thatconcern European policy. In this way the DGs and Serv-ices of the European Commission can access the worldof science and technology by utilising ISIS as the gate-way. As proposed in the Communication Towards a

European Research Area, networks will be developedto serve as common scientific and technical referencesystems for policy decision-making.

COLLABORATIONS

Collaborations and partnerships enable ISIS to providean efficient and fast service to its customers on a widespectrum of policy-based scientific and technologicalissues. Typical forms of collaboration include: shared-cost action collaborative projects under the FrameworkProgramme; institutional collaborations without ex-change of funds; sharing of experimental facilities;commercial services for third parties; sub-contracting;workshops; research networks; technology licences;and supplies and services contracted by the JRC. Par-ticipation in these collaborations enhances the scien-tific competence and scientific validity of ISIS, increasesits visibility and is a means of attracting new researchers.

INFRASTRUCTURES

ISIS operates a number of unique experimental facili-ties, for example, PERLA (Performance Laboratory)for safeguards and verification techniques, TEMPEST(Thermal, Electromagnetic and Physical EquipmentStress Testing) laboratory, ELSA (European Laboratoryfor Structural Assessment), LDTF (Large Dynamic TestFacility), LISA (Linear Synthetic Aperture Radar facili-ty), EMSL (European Microwave Signature Laboratory)and EGO (European Goniometric laboratory). As wellas positioning ISIS at the forefront of international re-search, they attract users and customers from MemberState and non-Member State laboratories, lend a visibleEuropean focus to research networks and provide themeans to address sensitive issues. These facilitiesrepresent unique research infrastructures in Europe.

TRAINING AND MOBILITY

ISIS encourages scientists from other organisations tospend time working at the JRC and contributes to theirscientific training and development by offering inter-disciplinary research in an international environmentwith a good infrastructure, some unique installationsand excellent networking opportunities. This has beenachieved through various mechanisms including MarieCurie fellowships and other Community and nationalmobility initiatives. ISIS is committed to continuing andstrengthening these arrangements and increasing theemphasis on hosting scientists from the Applicant Coun-tries so as to speed up their integration into the EU.

EXPLORATORY RESEARCH

ISIS currently earmarks 6% of its institutional budgetfor exploratory research. The aim of this mechanism isto test out ideas of a more speculative nature to assistISIS in shaping future actions. Exploratory research in-volves a bottom-up and top-down approach whereinprojects are proposed by the scientific staff of ISIS,evaluated in a documented and transparent mannerby a scientific committee on the basis of scientific in-novation, feasibility, strategic value and cost, and thenselected by a management committee comprising thedirector and unit heads.

In the period 1997-2000, 46 projects were approved.Nearly all yielded interesting results and peer-reviewedpublications. While several projects were developedfurther in collaborative partnerships with competitivefunds, others were amalgamated into the JRC’s institu-tional work programme either within ISIS or within otherinstitutes.

KNOWLEDGE MANAGEMENT

ISIS is highly aware of the importance of pooling itsskills and knowledge. At the level of the JRC, there is acorporate portal: a centralised home page with links tothe various institutes (http://www.jrc.cec.eu.int/isis)

and other items of information. At the level of ISIS,many Intranets exist that permit the storage of a coreof knowledge.

CUSTOMER RELATIONSHIPS

For each policy support theme in ISIS: (1) nuclear safe-guards, non-proliferation and international humani-tarian security, (2) cyber-security, knowledge accessand assessment, the fight against fraud and (3) natural

and technological risks and emergencies, a number ofcustomers can be identified. The main customers areidentified in the table below.

During 2000, ISIS sent out customer opinion question-naires to its institutional customers – various DGs andServices of the European Commission – in order tomeasure customer satisfaction and to evaluate ISIS’sresponsiveness to its customers. The questionnaire in-cluded several questions covering the usefulness ofthe research results, scientific quality, project man-agement, the administrative and financial interface ofISIS and pressing scientific or technical issues thatneed exploring for future policy needs. The responserate was 82 %. The results proved that ISIS offers aservice that meets the requirements of its customers.

7

ISIS’S MAIN CUSTOMERS FOR POLICY SUPPORT THEMES

Agriculture DG

Education and Culture DG

Energy and Transport DG

Enlargement DG

Enterprise DG

Environment DG

Euratom Safeguards O÷ce (ESO)

EuropeAid Co-operation Office (AIDCO)

European Anti-fraud O÷ce (OLAF)

European Environment Agency (EEA)

European Investment Bank (EIB)

Eurostat

External Relations DG

Fisheries DG

Health and Consumer Protection DG

Information Society DG

Internal Market DG

International Atomic Energy Agency (IAEA)

Regional Policy DG

Research DG

Secretariat General

Customers

Natural and technological risks and emergencies

Nuclear safeguards, non-proliferation and international humanitarian security

Cyber-security, knowledge access and assessment, the fight against fraud

8 Policy support themes

POLICY SUPPORT THEMES

ISIS’s scientific and technological base is designed tosupport EU policies related to the general theme of en-hancing the safety and security of the individual citizenand society. Participation in research, which is increas-ingly based on networks and collaborative projectswith industry, universities, national laboratories andresearch organisations, forms the backbone of thisbase. Research in ISIS is based on three main themes,which support European policy processes: (1) nuclearsafeguards, non-proliferation and international human-itarian security; 2) cyber-security, knowledge accessand assessment, the fight against fraud; and 3) naturaland technological risks and emergencies.

Nuclear Safeguards, Non-proliferation and International

Humanitarian Security: With its well-established anduniversally recognised role as a reference centre forthe technology of safeguarding fissile material and itsgrowing presence as a European focus for initiatives toprovide more efficient ways of reducing the threatfrom anti-personnel land mines, ISIS is able to providea solid support to the EU’s efforts to develop an effec-tive Common Foreign and Security Policy (CFSP). Thecustomers include the inspectorates (ESO – EuratomSafeguards Office and IAEA – International Atomic En-ergy Agency), the External Relations DG, the Develop-ment DG, AIDCO (EuropeAid Co-operation Office), ECHO(Humanitarian Aid Office) and the Information SocietyDG.

Cyber-security, Knowledge Access and Assessment,

the Fight against Fraud: The explosive growth of infor-mation and communication technology is bringing un-doubted benefits to society. There are, however, fearsabout the risks of fraud and threats to privacy, concernsthat a society that is increasingly dependent on globalcommunications might be vulnerable to systematicmalicious attacks on its infrastructure and worries thatcitizens who are poorly educated, disabled or mono-

lingual might be excluded from the benefits of theInformation Society. As well as supporting EU policyon these issues, ISIS has been supporting the EuropeanCommission’s efforts to increase the effectiveness ofanti-fraud measures. In addition to exploring the appl-ication of new technologies – DNA analysis for live-stock identification, intelligence gathering from opensources, language technology to analyse multilingualdocuments, satellite image interpretation – ISIS pro-vides customers with an integration of knowledge thatincludes the entire cycle from data capture, data fusion,data mining through to visualisation and estimation.Furthermore, ISIS has been working to provide moreaccurate, more timely and more socially robust infor-mation to the policy process, e.g., by improving officialstatistics and by developing quality assurance meth-odology for scientific input to governance. Support isgiven to the Information Society DG, the Education andCulture DG, the Agriculture DG, the Fisheries DG, OLAF,the European Environment Agency, EUROSTAT and theEuropean Investment Bank.

Natural and Technological Risks and Emergencies:

ISIS has developed a considerable expertise in themanagement of natural and technological risks and, ina number of important areas, is the main adviser tothose services of the European Commission responsiblefor managing these risks. ISIS’s work can be categorisedinto the following domains: harmonised European re-porting systems - e.g., Major Accident Hazards Bureau(MAHB), Natural and Environmental Disaster Informa-tion System (NEDIES) and the European Co-ordinationCentre for Aircraft mandatory Accident Reporting Sys-tems (ECCAIRS), the application of structural mechanicsto improve safety, transient structural dynamics andnuclear safety. Direct and impartial support is given tothe Environment DG, the Regional Policy DG and theEnterprise DG.

NUCLEAR SAFEGUARDS, NON-PROLIFERATIONAND INTERNATIONAL HUMANITARIAN SECURITY

1 Safeguards and Prevention of Proliferation of Nuclear Weapons 10

1.1 Mass and Volume Measurement Techniques (TAME Laboratory) 10

1.2 Non-Destructive Assay (NDA) 11

1.2.1 Improvements in PERLA 111.2.2Neutron Measurement Techniques and Calorimetry 121.2.3 Gamma Spectroscopy 141.2.4Unattended Verification of Fresh Fuel Assemblies using UMS 14

1.3 Sealing and Identification Techniques 15

1.4 Transponder-based Identification 16

1.5 Surveillance and Monitoring 17

1.5.1 Design Information Verification 171.5.2 Image Retrieval by Visual Examples 171.5.3 Machine Learning Techniques for Data Reduction of Surveillance Images 181.5.4 Visualisation Techniques for Training of Inspectors 181.5.5 Electronic Agenda 181.5.6 Remote Inspection and Tele-operation for Advanced Storage Areas 19

1.6 Environmental Testing (TEMPEST laboratory) 19

1.7 Information Systems 20

1.7.1 Satellite Imagery and Geographical Information Systems 201.7.2 Information Centre on Non-proliferation 21

1.8 Inspector Training 21

1.9 Technical Assistance to the Commonwealth of Independent States (TACIS) 22

1.9.1 RMTC (Obninsk) and URSIMTC (Snezhinsk) 221.9.2Instrument development 231.9.3 NMAC at the ULBA facility 23

1.10 ESARDA 23

2 Humanitarian Demining 24

2.1 Improving Transparency, Visibility and Effectiveness of EU Mine Actions 24

2.2 Remotely Controlled Mechanical Equipment 25

2.3 Pilot CAD Database for the Close-in Detection of Landmines 25

2.4 Human Factors in Humanitarian Demining 25

Nuclear Safeguards, Non-proliferation and International Humanitarian Security 9

OVERVIEWA. PoucetThe mission of ISIS in the safeguards and non-proliferation area is to provide enabling research,technology, instruments, technical services and train-ing for the verification of treaties relating to the non-proliferation of weapons of mass destruction:

• to Inspection Agencies;• to States and operators establishing and operating

accountancy and control systems;• as a support for the definition of a common EU non-

proliferation policy.

Verification and control related to non-proliferation ofnuclear materials is performed by the Euratom Safe-guards Office (ESO) of the European Commission withinthe context of the Euratom treaty, and by the Interna-tional Atomic Energy Agency (IAEA) within the contextof the Non-Proliferation treaty. ISIS gives scientific,technical and training support to both these organisa-tions.

ISIS also supports the Russian Federation and otherCommonwealth of Independent State (CIS) countriesin establishing a system of nuclear materials account-ancy and control, which is comparable with Westernstandards in terms of efficiency and effectiveness. Inaddition to states and agencies, ISIS also helps facilityoperators to meet their safeguards obligations in anefficient and cost-effective way.

To be able to provide state-of-the-art support to itscustomers, ISIS carries out underpinning research anddevelopment in a number of areas that are relevant tonuclear materials accountancy and control and to otherverification activities in the framework of non-prolife-ration of weapons of mass destruction. Such researchis necessary to meet the demands for more cost efficientsafeguards and to respond to new emerging needs dueto changes in verification protocols in nuclear fuel cycletechnology and materials.

1.1 MASS AND VOLUME MEASUREMENTTECHNIQUES (TAME LABORATORY)B. Hunt

Safeguards support activities have been performedfor the IAEA and ESO, and for the establishment ofmass/volume and containment/surveillance infra-structure and training at the ULBA fuel fabricationplant in Kazakstan (under the TACIS programme).

For the Joint Members Support Programme task to theIAEA (Independent Data Collection and Evaluation dur-ing the LASTAC Exercise) a joint draft document wasproduced on the conclusions and recommendations,

which will be submitted to the JNFL (Japanese NuclearFuel Limited) by the IAEA [1.1]. The document concernsa safeguards system for the Rokkasho ReprocessingPlant in Japan. The joint members completed the drafton target for submission in October 2000 followingmeetings in January and September 2000 on the reviewof analyses and recommendations resulting from theexperimental campaigns of 1999. IAEA personnel alsoparticipated in a mass/volume methodology trainingcourse in November 2000 and received special instruc-tion on the JRC-ISIS’s Unattended Volume Measure-ment System (UVMS), which incidentally has beenpurchased by NMCC (Nuclear Materials Control Centre)of Japan and scheduled for delivery after ISIS verifica-tion, assembly, commissioning and test reporting inDecember 2000, with a view to application in Rokkashoin June 2001. Similarly, requests were received in Dec-ember from the IAEA for the purchase of two UVMSsystems for inspector application in the Rokkashofacility in 2001. This is currently being processed.

In the field of R&D, long-term comparison experimentshave been and are continuing on both the PIMM (Pres-sure Instrument Measurement Module) prototype instr-ument and the commercially available digital pressuremodules DPM 7885. Tank calibrations, long-term mon-itoring and sensitivity tests have been performed andare undergoing analyses [1.2].

For ESO support programme tasks, the planning madelate 1999 with respect to the wishes of ESO regardingtank calibrations in La Hague, France was successfullycarried out during 2000. In particular equipment wassupplied and commissioned at the JRC prior to ship-ment to La Hague. Up to five man weeks of supporton-site was also provided by ISIS personnel in July andAugust, in setting up the equipment, commiss-ioning itand launching the first calibration campaigns, withESO personnel in attendance for on the spot trainingin order to be able to continue with the rest of thecampaign. The equipment is due to be shipped back toIspra in January 2001 after the completion of the finaltank calibration. Two training courses were carried outin both March and May 2000 in the Mass/Volume labor-atory for ESO inspectors closely involved in the LaHague campaigns. Experience with the equipmentdestined to be utilized in La Hague was also includedas part of the training course.

Regarding the TACIS work, an administrative arrange-ment [1.3] was set up with respect to support activities forthe ULBA conversion and fabrication facility in Kazak-stan. On the basis of an analysis made on the situationin the ULBA conversion and fabrication lines, a prioritywas established to improve the following techniques:

10

Safeguards and Prevention ofProliferation of Nuclear Weapons1

• Mass/Volume determination of liquids in a largenumber of tanks (M/V techniques).

• Procedures for calibration and data evaluation forweighing scales (M/V techniques).

• Sealing and identification techniques (C/Stechniques).

• Surveillance techniques (C/S techniques).

A visit to the ULBA facility in June 2000 allowed firsthand knowledge and technical information of the plantto be gathered. Sub-contracts have been set up withthe ULBA Metallurgical Plant and the project was de-fined with the aid of the Swedish consultancy firm, SKI.A detailed analysis of the plant [1.4] has been madeallowing a clearer definition of priorities. Preparationof the training room will have priority together withtraining courses in Ispra and subsequently in the ULBAplant itself. For the mass/volume determinations, thedesign prototype of a fixed measurement station incor-porating a micro-scanning valve system and a porta-ble measurement station together with the informaticsfor data acquisition and management have been definedwith components already purchased and under test.Similarly for containment surveillance, equipment hasbeen purchased or ordered, such as hardware for theverification of metal cap seals, a market survey for theembedded computers in the camera housing of thesurveillance system, plus the development of a soft-ware module for ambient tests of the embedded com-puter with TV. The latter is extremely important due tothe yearly extremes of temperature, -50 ºC to +40 ºC.Preparation is underway for the technical specificationsregarding the tender which forms 50% of the TACIScontract for equipment purchase.

For the TACIS 98 budget relating to Kazakstan, thetechnical terms of reference are already underwaywith the completion of the administrative arrangementexpected early 2001.

1.2 NON-DESTRUCTIVE ASSAY

1.2.1 Improvements in PERLA

P. SchillebeeckxThe nuclear safeguards activities, related to Non-Des-tructive Assay (NDA) of nuclear materials, are executedin the PERformance LAboratory (PERLA). PERLA is aunique facility within the EU, housing an extensive col-lection of well-characterised nuclear reference materialand instrumentation. The activities are specifically ori-ented to the development, validation and implementa-tion of nuclear NDA methods and techniques. The app-lication fields cover calorimetry, gamma spectrometry,passive neutron assay and active neutron interrogation,used either individually or in integrated systems. Mostof the activities are done in direct support to ESO andthe IAEA. Over the last year the laboratory performancewas significantly improved, with a view to accreditationas a European reference laboratory for NDA of nuclearmaterial. Already during 2000 these improvementshad a direct impact on the support to national and inter-national nuclear safeguards authorities and the nuclearindustry.

The performance of PERLA is based on three pillars:nuclear reference material, reference instrumentationand the experienced staff developing various method-ologies for the NDA of nuclear material. The procurementscheme and characterisation of reference neutronsources, and the procedures for characterisation ofneutron instrumentation were improved. This resultedin an enhanced use of the Monte Carlo simulation codeMCNP-PTA and the neutron correlation techniquebased on Hage’s point model developed at ISIS.

References

1.1 Authors from IAEA, USDOE, NMCC Japan and JRC-Ispra.

Joint MSSP Task A1132, “Independent Data Collection andEvaluation during the LASTAC Exercise, – Conclusionsand Recommendations”. Safeguards Confidential.

1.2 Landat D., Caviglia M. – Tests on PIMM3-M & DPM7885, Calibration, Monitoring and Sensitivity –November 2000. Final document in preparation.

1.3 Administrative Arrangement No. 15945-2000-04 A1COISP BE, Project ULBA/TACIS97 – K.5.01/97, “Establish-ment of Facilities for Mass/Volume, Containment/Sur-veillance and Training at the ULBA Fuel FabricationPlant in Kazakstan”.

1.4 Landat D. – Mass/Volume Measurements Tacis K5.01.97“Preparation of the Technical Specifications” –November 2000. Final document in preparation.

1.1 Mass and volume measurement training session in the

TAME laboratory.

11Nuclear Safeguards, Non-proliferation and International Humanitarian Security

1.2 PERLA and its functions.

• Characterisation of neutron sources

The characterisation and calibration procedures ofneutron sources were updated and include neutron,gamma and heat measurements. This results in a det-ermination of the total neutron emission rate, totalactivity and nuclide composition. The total neutronemission rate is obtained from measurements withneutron counters optimised for application of the neu-tron correlation technique. The counters are character-ised and calibrated by both Monte Carlo simulationsand measurements with reference neutron sources.The sources are certified for their neutron intensityand anisotropy by the National Physics Laboratory inthe UK. Application of the neutron correlation techniqueallows the absolute determination of the emission ratefor 252Cf(s.f.) sources and the verification of the degreeof expected randomness for the (α,n) isotopic neutronsources. For (α,n) sources the total 241Am activity isdeduced from measurements with an Anti-Comptonspectrometer surrounding a Ge-telescope. This spec-trometer is characterised by Monte Carlo simulationsand calibrated by three independent laboratories. Fromthe photon spectra the relative abundance of impurities(e.g. 154Eu, 239Np, 243Am,…) is deduced. In addition theintensity of some Doppler broadened full energy photonpeaks indicate the probability of (α,n) reactions andcan be related to the neutron output. The total activityof the (α,n) sources is also determined by calorimetricmeasurements using the Small Sample Calorimeter(SSCAL). This calorimeter has an optimum operationregion of between 5 – 80 mW and a 0.15% measure-ment precision.

• Methodology for calibration of neutron devices

used for NDA

The PERLA procedures for the characterisation andcalibration of instruments, used for passive and activeneutron assay, have been improved to include the useof new reference instrumentation and certified refer-ence material and the use of our Monte Carlo simula-

tion code MCNP-PTA. The complete process is opti-mised to allow a reliable determination of responsefunctions by MCNP-PTA and a verification of the instru-mental constraints for the application of Hage’s pointmodel, for the application of passive neutron assay.The complete characterisation process results in thedetermination of the following instrumental parame-ters: high voltage plateau and optimum operating bias,system dead time, empirical dead time coefficients,detection efficiency for a 252Cf and a plutonium “pointsource”, detection efficiency for various (α,n) neutronsources, radial and axial efficiency profile, and decaycurve for a point source in different positions of thecavity.

Calibration measurements of PERLA NDA standardsallow a comparison with the results of Monte Carlosimulations and/or the response of a reference instru-ment. For passive neutron instruments, the procedureincludes measurements of a set of plutonium stand-ards with a plutonium mass in the specific applicationrange of the instruments. For active devices, used forthe verification of fuel assemblies, the calibration reportincludes the response for typical BWR and PWR refer-ence assemblies and MTR assemblies if required.

1.2.2 Neutron Measurement Techniques and

Calorimetry

P. Schillebeeckx, M. ThorntonActive neutron interrogation is the most commonlyapplied Non-Destructive Assay (NDA) technique forthe determination of the fissile content in uraniumbearing samples. The verification of input powder andscrap material by ESO in Low Enriched Uranium (LEU)facilities is based largely on measurements with thePhonid device. The disadvantages of this instrumentare the need for very extensive and expensive calibra-tion campaigns and an interrogating neutron source.Together with the support group of ESO, a new meas-urement method, based on the detection of neutronsemitted after the spontaneous fission of 238U, is pro-posed. To investigate this technique, measurementswith a commercially Active Well Coincidence Counter(AWCC) were performed in PERLA. Besides thesemeasurements, various simulations using the MCNP-PTA code have commenced to confirm the proposedmethodology and to design a device, optimised for in-field use in LEU fabrication facilities of the EU.

The Physical Inventory Verification (PIV) in LEU facilitiesduring 2000 again indicated shortcomings for meas-urements on fuel assemblies with gadolinium loadingscurrently encountered. The calibration procedures forthe verification of such assemblies should be re-eval-uated. A re-evaluation, only based on experimentalresults, requires an enormous number of reference

12

M

ethodolog

y

Nuclear M

aterialInstrumentation

ReferenceCalibration andperformance evaluation

Training

Waste andDecommissioning

Development ofintegrated systems

In field solutionsPhonid ReplacementLEU Fuel AssembliesRainbow Project

Laboratoryfor

Safeguards and VerificationTechniques

Users and Clients:

ESOIAEA

TACISIPSN

Industry

materials and measurement campaigns, and is there-fore very expensive and time-consuming. To overcomethis problem the MCNP-PTA code, applied in combina-tion with neutron devices characterised according tothe methodology outlined above, was used. After ananalysis of the measurement data with the MCNP-PTAcode at ISIS, the operator’s declaration could be veri-fied. To avoid such interventions an analysis method-ology, always based on MCNP-PTA, is being studied incollaboration with the ESO.

• Support to ESO

The Drum Monitor, designed and manufactured at theJRC-Ispra was successfully installed in the old SiemensMOX fuel fabrication plant in Hanau (Germany). TheDrum Monitor is intended for the verification of Sie-mens’ declarations of the plutonium content in wastedrums. The instrument will be operated jointly by ESOand IAEA inspectors. The plant, now known under thename of “Siemens AG Rückbauprojekte”, will over thenext 24-36 months produce the last three thousand ofa total of about eight thousand 220 litre radioactivewaste drums. The content of all waste drums will even-tually be compacted, cemented and stored in a newstorage facility on site. At this moment about half of alldrums has been cemented. The waste includes compo-nents such as glove boxes and machine componentsused in the MOX fuel production as well as processwaste such as clothing and gloves. The drums (ce-mented and not cemented) may contain from belowone gram to about 30 grams of plutonium. Actinidesother than plutonium present in the waste are uraniumand americium in variable amounts. The operator gavethe permission for ESO (and ISIS) to access the DrumMonitor computer remotely via a dedicated telephoneline. The significance of this feature was widely appre-ciated by ESO as a means to reduce the number ofinspections to the facility. For ISIS this allows remotemonitoring of the system and quality checking of themeasurements.

A new procedure for measurements of so-called “Rain-bow cans”, which are difficult to characterise due tothe heterogeneous nature of the plutonium within thecan, has been validated in PERLA. Measurements ofsimulated rainbow cans were performed in PERLA usingplutonium standards of mixed isotopic composition,a gamma spectroscopy system and a calorimeter.An average isotopic composition was determined byrotating the cans, and moving the detector a sufficientdistance from the source, to allow the whole can to beviewed. Off-line post analysis of collected calorimetrydata enabled an improved prediction algorithm to bedetermined. Utilising this method enables measure-ment times to be reduced from in excess of 8 hours to 3.Following the successful completion of the tests a new

measurement methodology is being recommended toESO.

• Support to IAEA - verification of special nuclear

material entering the civil fuel cycle

As part of a collaboration agreement with the IAEA, aseries of measurements were performed to demon-strate the feasibility of using calorimetry in place ofdestructive analysis in the homogeneity testing of plu-tonium oxide from excess weapons material stored inthe USA. The PERLA Small Sample Calorimeter (SSCAL)and nuclear reference material were used to validatethe method prior to implementation in the field. Theprecision, accuracy and linearity as a function of heatoutput were verified by a set of heat producing sam-ples that both covered and exceeded the heat range ofthe instru-ment. If calorimetry is to reduce the burdenof destructive analysis, the final overall uncertainty onthe heat output must be less than 0.2%. Both electri-cal heat and certified plutonium standards have beenused in the performance evaluation of the instrument.Although not certified for heat output, the plutonium issufficiently well characterised to allow for its use as acalibration standard. The precision obtained with boththe electrical and plutonium standards is better than0.15% at 10 mW and, as such, exceeds the specifica-tion requested. It is envisaged that a reduction in thelevel of precision will be obtained.

• Support to the Institut de Protection et de Sûreté

Nucléaire (IPSN)

The improved quality procedures for the characterisationof neutron sources and NDA instrumentation resultedin a close collaboration with the Institut de Protectionet de Sûreté Nucléaire (IPSN), the French nationalsafeguards authority. During 2000 the first agreementhas been made for several calibration campaigns inPERLA during 2001. IPSN will also use PERLA as theEuropean independent reference laboratory to evalu-ate the performance of commercial instrumentationfor the measurement of plutonium samples prior toprocurement.

• Support to the nuclear industry

Further to the IAEA task related to the performancecomparison of codes for the determination of the iso-topic composition, much interest has been generatedwithin the nuclear industry. For example the gammaspectra library, created within the IAEA task activities,is used frequently by companies and international org-anisations as part of their quality control. At the requestof several organisations it is planned to make the libraryavailable on the Internet.


1.2.3 Gamma Spectroscopy

R. BerndtCdZnTe semiconductor detectors for gamma spec-trometry have an increasing importance as a measure-ment tool for Euratom inspectors in the field of nuclearsafeguards. Since they are small and handy they are aswell helpful in the fight against illicit trafficking of nuc-lear or radioactive material. This rapidly developingdetector type has found an application as a qualitativemeans for underwater measurements on spent nuclearfuel. Here, fission products or activation products needto be identified. A user code BE-VERIF has been devel-oped [1.5].

On the other hand, this detector has shown that it hasthe potential to give also good quantitative resultswith the characterisation of the isotopic compositionof radioactive samples. There is still however a draw-back to overcome: the spectral responses of all indi-vidual detector specimens vary so strongly - (Figure

1.3) - that a uniform concept for the peak area determi-nation results in errors of a factor of two. For this rea-son the individual characterisation of detectors wasstarted according to a peak shape model with fourcomponents. Using the detector specific and energydependent parameters, the uncertainties are dramati-cally reduced and it is possible to determine isotopicratios in a reliable way. This work is aimed at usertools in the above mentioned fields.

1.2.4 Unattended Verification of Fresh Fuel Assemblies

using UMS

P. Schillebeeckx, J. Löschner, J.G.M. GonçalvesUnder the support programme to ESO, ISIS has devel-oped and tested unattended automated systems for

the safeguards verification of fresh LEU and MOXassemblies in fabrication plants. The implementationof such systems will allow ESO to have 100% verifica-tion of fuel assemblies without inspector presence.The unattended verification includes video reading ofthe assembly identification number and a neutron col-lar measurement. The position and duration of themeasurement can be configured according to the ass-embly type if the identity number of the assembly isknown to the system. If the assembly characteristicsare known to the system, it can carry out unattendeddata evaluation to confirm the declared characteris-tics. The verification is activated by the operator load-ing the assembly into the measurement station and noother intervention of the operator is required. The un-attended measurement station (UMS) carries out itsown monitoring of background and detection efficiency.

During 1999, the UMS-2 was extensively tested at theABB LEU fabrication facility at Västerås (Sweden). Thiswas carried out under a collaboration contract betweenABB and ESO. The objectives of the test were to see theimpact of this safeguards approach on the productionflow in real manufacturing conditions and to verify theprecision and accuracy of the measurements. Thistesting period gave very satisfactory results. One fea-ture of the test was that during operations both meas-urement data and operational status information couldbe accessed using a remote link to the ISIS laboratories.In the last quarter of 1999, the UMS-2 was used by ABBduring production of three different types of PWR ass-emblies and the resulting experience was evaluatedby ISIS, ESO and ABB. The data from these measure-ments were analysed by ISIS in order to quantify themeasurement performance of the instrument.

In Figure 1.4, the UMS-2 measurement uncertainty formeasurement campaigns of two different fuel assemblytypes is given. The data, originating from assemblieswith and without neutron poison, show that the repro-ducibility is slightly more than 1 % (without countingstatistical uncertainty).

14

1.3 Typical variations of the peak shape of good CdZnTe

gamma radiation detectors.

1.0

0.8

0.6

0.4

0.2

0.01150.0 1200.0 1250.0 1300.0

ENERGY / keV

COU

NTS

/ re

l. un

its

References

1.5 Braun A. – Diplomarbeit – Ispra, 2000.

1.4 Illustration of the UMS-2 measurement uncertainty.

0.04

0.02

0.00

-0.02

-0.040 5 10 15 20 25 30 35

Assembly Identifier

Rel

ativ

e D

iffer

ence

Type 1 without Gd

Type 2 with 12 Gd Rods

During 2000, discussions have taken place betweenESO and the FBFC LEU fabrication facility in France (Ro-mans sur Isère) to prepare the installation of two UMSsystems on the production lines, which are being setup there. It is planned that the first of these linesshould be functional in 2002 and the second in 2005.The first UMS system needs to be ready for installationin August 2002. Technical discussions involving FBFC,ESO and ISIS are ongoing in order to identify the tech-nical modifications required for the features of thespecific production lines. In this context, ISIS is inves-tigating the applicability of triangulation-based laserstriping for reading the characters that identify the ass-embly while it is in the UMS. ISIS is also investigatingthe integration of a vision-telephone system (web-based) so as to allow ESO inspectors have access toremote ISIS technical support during inspector visitsto the UMS. The neutron collar measurement approachwill remain as in the earlier versions.

1.3 SEALING AND IDENTIFICATION TECHNIQUES(CONTAINMENT SURVEILLANCE)B.C. d’Agraives

The sealing and identification techniques laboratory(SILab) activities have essentially been an extension ofthe work done in 1999. SILab has been active in threedomains and has progressed mainly in testing and in-stalling new equipment to meet research needs. In ad-dition, an important third party work contract has beenadvanced. SILab has specialised in developing ultra-sonic sealing techniques as a support to, and on requestof, the safeguards authorities such as the IAEA or ESO,which originally requested such systems for the con-trol of nuclear storage containers, either in La HagueCogema plant/France (MK-6 Sealing-Nuts) or in Sell-afield BNFL plant/UK (MK-4 Sealing-Bolts).

In 2000, the activities focused around three main lines:

• Support and Assistance: to ESO and, to a lesser ex-tent, to the IAEA, which jointly or separately use SILabsealing systems on storage containers at the abovementioned sites. The SILab team visited La Hague(France) and Sellafield (UK) sites several times andprovided support to inspectors as well as replacementequipment. New orders for reading equipment andseals were prepared and delivered as well.

• Research Activities: work continued on the im-provement of the sealing bolts’ mechanism to offer asafer item to inspectors. A large effort was also devotedto the specification and fabrication of special electroniccards to be installed on portable computers used forthe verification of seals. On a principle similar to thatfor ultrasonic seals, SILab has also studied, developed

and patented a new technique for the “unique iden-tification” of sensitive structures such as PWR fuelbundles or any other kind of container, by incorporatingspecially marked metallic “disks” generating a digitalcode to be “uniquely” read by ultrasonic means. In thearea of developing identification means, SILab has alsostudied, developed and filed a patent for a new “non-ultrasonic” technology, based on laser interferometry,which is aimed at the identification of sensitive itemsby using “speckle images” produced in certain condi-tions in a reproducible manner by their surface (Figure

1.5). SILab has worked and solved the main mechanicaland electronic problems raised by a special “doubleway” ultrasonic reading-head, capable of reading twoultrasonic features (identity and integrity) or (identityand number) practically at once.


1.6 Production of parts for seals for MOX fuel bundles before

assembly and laser welding.

1.5 New interferometric head for the identification

of objects by “Speckle Effect”.

• External collaboration: SILab has conducted twokinds of external collaboration: (1) manufactured tworeading units associated with two portable computers(which were to be used for demonstration purposes)and provided them to the TACIS programme, along withsome ultrasonic seals specimens for the IPPE ObninskCentre in Russia and (2) worked intensively on the230,000 Euros contract signed last year with the Japa-nese Hitachi/Toshiba Fuel Makers (now JNF). This con-tract aims at offering a feasible solution to JNF fuelmakers if they have to place ultrasonic seals on theirfuture BWR MOX fuel bundles, which should come intoproduction/be used in Japan within the next years.This work is based on an original skill, developedyears ago, when SILab developed and tested a special“cap seal” for German BWR fuel bundles in Kahl (Ger-many). Several technical visits were made to Yokohamaand Ispra. SILAb came up with 5 different designs andseries of prototypes, which were all assembled andtested mechanically (Figures 1.6 and 1.7). If the contractis not extended, it will end in March 2001.

1.4 TRANSPONDER-BASED IDENTIFICATIONC. Korn

The transponder-based seal was further developed in2000 to eliminate the need for an inspector to tie aknot in the wire. This new wire-locking system reducesinstallation problems and is less time-consuming.3000 samples of this ISIS patented transponder seal(Figure 1.8) were delivered to the ESO for installationand evaluation. New readers for in-field identificationare under evaluation in the TEMPEST laboratory – seeSection 1.6. These readers combine a high readingrange with a larger memory and better overall resist-ance.

Another batch of 300 seals has been installed in theIspra decommissioning building to identify various typesof containers. Special disk transponders have also beeninstalled on the transport pallets for identification onentry to and exit from the various storage facilities.Specially designed stationary readers were studiedand developed and are being installed to track themovements of the containers (Figures 1.9 and 1.10).

16

1.7 Laser welding of seals for Japanese MOX fuel bundles.

1.8 View of the new PTS seal with wire locking system.

1.9 View of the antenna and a palette with disk transponder.

1.10 View of the containers with their individual tags.

1.5 SURVEILLANCE AND MONITORING1.5.1 Design Information Verification

V. Sequeira, J.G.M. Gonçalves, E. WolfartThe objectives of design information verification (DIV)are threefold:• to ensure that an installation has been built accord-

ing to the approved design plans;• to ensure that an installation has not been changed

without authorisation;• to track and document all the changes in successive

inspections (continuity of knowledge).

The process of DIV can be divided into two main steps:acquire raw data from the plant and compare the “as-built” and target models. The objective of this work isto develop a prototype that accepts multi-sensory dataas input and is multi-scale. This last feature allows fordifferent acquisition systems and algorithms accord-ing to the size of the objects/buildings to be modelled.Indeed, a building, large tanks or small pipes requiredifferent equipment and modelling algorithms. Thesystem will provide semi-automated tools to comparethe acquired “as-built” and the approved design models.Figure 1.11 shows a short-range laser scanner as wellas a snapshot of the 3D model of a real pipe scene.Safeguards relevant distance measurements can beautomatically extracted.

1.5.2 Image Retrieval by Visual Examples

E. Stringa, J.G.M. Gonçalves, P. MeylemansSurveillance cameras installed in nuclear plants pro-vide valuable safeguards information. A major problemis the number of images to be analysed by inspectors.A first reduction, typically 95%, can be obtained by usingscene-change detectors, but this is still not enough.The goal of this research is to design and develop asystem capable of identifying all safeguards relevantimages as a support to inspectors in the analysis ofsurveillance images.

Image retrieval systems aim at extracting from an imagedatabase all the images similar to a specific, user-defined example. This is achieved by characterisingeach image (i.e., indexing) with a set of numerical fea-tures (related to shape, colour, texture and other prop-erties). The retrieval phase uses a subset of features torank the images by decreasing degree of similarity. Theuser queries the database by selecting a visual exam-ple (an example image) and the interesting picturesare extracted by matching the content of the exampleto one of the images in the database (Figure 1.12). Pre-liminary tests indicate the suitability of image retrievaltechniques [1.6-1.10] to extracting all safeguard relevantimages from a surveillance database.


1.11 A laser range scanner measures the real pipe scene “as-built” and provides dimensionally accurate data for the 3D modelling

and verification process.

1.12 The user queries the database by selecting a visual example (an example image) and the interesting pictures are extracted by

matching the content of the example to one of the images in the database.

Image database Retrieval engine

Feature extractionFeature extraction

Example image

Retrieved imagesDatabase creation (indexing)

Query

S=100% S=96.643% S=94.511%

S=93.761% S=92.294% S=91.877%

1.5.3 Machine Learning Techniques for Data Reduction

of Surveillance Images

J. del R. Millán, M. Franzé, J. Mouriño, C. VersinoIn available image retrieval systems, the user mustmanually select those numerical features that betterdescribe the relevant images. This work explores machinelearning techniques for the automatic selection andtuning of relevant features to retrieve safeguard relevantimages.

The result of a preliminary study indicates that a prom-ising approach is to combine feature selection and induc-tion – i.e., the process of learning the appropriate class-ifier that differentiates between safeguards relevantand irrelevant images – in an interactive system wherethe inspector gives qualitative feedback on a limitednumber of images. Briefly, by giving two initial sets ofrelevant and irrelevant images, the system selects thosefeatures for which the classifier better discriminatesbetween the two classes. Then, the system applies thelearned classifier over new images and presents a fewsignificant images to the inspector for feedback wholabels them as relevant or irrelevant. After a few itera-tions the initial sets of images expand and the learningof the appropriate classifiers is tuned gradually.

1.5.4 Visualisation Techniques for

Training of Inspectors

J. del R. Millán, M. Franzé, J. Mouriño, C. VersinoAdvanced multimedia and visualisation techniquesprovide a major contribution to the realism required inany safeguards training package, considering that anobject (e.g., an instrument) or remote location (e.g.,nuclear plant) is represented by different types of data:text, photographs, animations, videos, sound, 3D, etc.(Figure 1.13).

A visual panorama is a 360˚ Internet compatible pictureobtained by stitching together a sequence of adjacentphotos. Several visual panoramas can be connected tocreate a virtual tour – an interactive visit to all relevantlocations of a plant. This provides a good perception ofspace, i.e., the “feeling of being there”. Specific regions,i.e., “hot-spots”, can be easily linked to other sourcesof information. It is thus possible to link a large instal-lation to its contents, inventory information or detailedexplanations on how to use/maintain the equipment.Recently, these techniques have been demonstratedfor the creation of safeguards training tools.

1.5.5 Electronic Agenda

V. Sequeira, S. Carelli, C. Korn, J.G.M. GonçalvesRecent developments in information and communicationtechnologies are changing the way professionals work.This is particularly true for people working away fromheadquarters, as is the case for safeguards inspectors.Three technologies are instrumental in this change:• Miniaturised computer equipment (e.g., palmtops,

pocket PCs or wearable devices);• Wireless communications both short (e.g., Bluetooth

or IEEE 802.11b) and long range (e.g., GSM);• Position technologies (e.g., GPS or transponder-

based identification).

The combination of the above technologies enablesthe development of a new family of field equipment tosupport safeguards inspectors when away from head-quarters. The objective of this work is to identify themost suitable technologies, which allow for easy andsecure remote access to safeguards information (e.g.,manuals, maintenance instructions, measurement data-bases, etc.). In a second phase some working proto-

18

Suggested Further Reading

Versino C., Millán J. del R., Franzé M., Mouriño J.

Machine Learning Techniques for Data Reduction ofSurveillance Images – Internal report 2000.

References

1.6 COBWEB:http://cobweb.kpnqwest.no/summary.html

1.7 COMPASS:http://compass.itc.it/index.html

1.8 QBIC:http://wwwqbic.almaden.ibm.com/

1.9 SURFIMAGE:http://www-rocq.inria.fr/cgi-bin/imedia/surfimage.cgi/

1.10 IMATCH:http://www.mwlabs.de/

1.13 Internet-compatible visual panorama tour of the European

Commission’s building “Cube” at Luxembourg. The user

“walks-through” the building either by clicking on a

specific room or by following the pre-defined spatial links.


types based on commercial off-the-shelf componentswill be built to test the validity and usability of the newequipment. Figure 1.14 shows a subset of today’s com-ponents that can be already employed by safeguardsinspectors in the field.

1.5.6 Remote Inspection and Tele-operation for

Advanced Storage Areas

E. Ruiz MoralesAfter the dismantling of nuclear weapons and theirprobable release under international verification regi-mes, unmanned, highly automated and secured envi-ronments may be required for the storage of strategicmaterials. In view of this technological emergence,ISIS is developing an advanced prototype and demon-stration system for the automated control of storageareas. This system also aims at being a training plat-form for inspectors and a test-bed for TACIS projects innuclear safeguards.

The RIALTO tele-operation laboratory includes a proto-type robotics system, and associated tools such as theGENERIS robotics controller, the graphical computersimulation for real-time monitoring and the inspector’sinterface. Further, a variety of sensors were investigatedin order to provide autonomy to the robotics equipmentduring the material inspection phases. As a result, avision system and tactile force/torque sensors weresuccessfully integrated (Figure 1.15). In addition, thedesign of a novel tele-operation concept including amobile platform and special guidance equipment forenhanced flexibility was recently initiated

1.6 ENVIRONMENTAL TESTING(TEMPEST LABORATORY)C. Korn

The TEMPEST (Thermal, ElectroMagnetic and PhysicalEquipment Stress Testing) laboratory extended its ac-tivities in terms of electromagnetic compatibility im-munity and measurement testing (such as harmonicsmeasurements, conducted perturbations measure-ments, etc.) as well as in the field of mechanical shocksand low frequency vibrations, installing new generatorsand measurement instruments to be able to performtests based on new international standards.

The accreditation of the TEMPEST laboratory accordingto quality standard EN 45001 was confirmed in Decem-ber 2000 with an external COFRAC (COmité FRançaisd’ACcréditation) surveillance audit and a new test benchwas recognised: immunity of conducted disturbancesinduced by radio-frequency fields (according to IEC1000-4-6 standard). The TEMPEST laboratory remainsthe only thermal and electromagnetic test laboratory,which is recognised world-wide under the EN 45001quality standard.

During 2000, the activities of the TEMPEST laboratoryincreased in terms of the number of safeguards equip-ment and devices tested mainly in support to the IAEA,for example, electromagnetic (radiated and conductedimmunity and emission) tests conducted on IAEA’sDCM-M3 [1.11], ALIS Pacpro [1.12] and Portable Alis[1.13] surveillance systems, as well as thermal, humidityand mechanical tests conducted on the portable readerfor the JRC passive transponder seal [1.14] used by theESO as well as on the IAEA’s Alis Pacpro [1.15] surveil-lance system.

AccreditationNo. 1-1037

1.14 Example of commercial off-the-shelf components for an

Information Agenda to be used by safeguards inspectors

in-field.

PCMCIA Wireless

Pocket PC

Ethernet Link

Access Point

Transponders Antenna

Reading StationRS232

1.15 Handling of a plutonium container mock-up for inspection

tasks by a remotely controlled robot.

In order to be able to match the laboratory tests asclosely as possible to the real equipment operationconditions, the TEMPEST laboratory, as a support tothe ESO, developed a specific portable environmentalparameters recording unit. (Figure 1.16). This recordingunit, transported and installed where ESO surveillancesystems and/or seals and measurement instrumentsare located, is able to record the following parameters:temperature, humidity, light intensity, radiated electricfields, radiated magnetic fields and main network per-turbations (voltage variations, harmonics, surge etc.).A special training course was held in the TEMPEST lab-oratory for ESO experts in order to manage the system,which will be used to correlate in the best possibleway in-field operation conditions of ESO equipmentwith the test levels conducted in the TEMPEST labora-tory. This will result in the development of a new testprocedure for safeguards equipment, based on in-fieldmeasurements and IEC standards.

The TEMPEST laboratory began a new activity of supportto the Energy and Transport DG within the frameworkof the new recording equipment for road transport.The role of the TEMPEST laboratory is defined as follows:• development of specific test procedures and test

benches for interoperability tests between on-boardcontrol equipment, sensors and tachograph cards;

• interoperability tests;• certification of public keys for the security of the

systems.

The TEMPEST laboratory has prepared the design of thevarious test benches as well as the overall organisation.Preliminary tests will be run in 2001.

1.7 INFORMATION SYSTEMS1.7.1 Satellite Imagery and Geographical

Information Systems

A. PoucetISIS started to investigate the benefits and cost of usinghigh-resolution satellite images as a means for providingindependent information on layout and characteristicsof nuclear sites. High-resolution satellite images, withground sampling distance of a few metres – until recentlythe domain of military satellites – are now becomingavailable on a commercial basis and can provide usefulinformation to verify site declarations.

It was realised early on that a promising approach wasto integrate satellite images with other geo-referencedinformation (Figure 1.17). Therefore a start was madewith the development of an “inspector workstation”based on Geographical Information Systems (GIS) tech-nology. In this approach satellite pictures are corrobo-rated with other information such as site maps and othergeographically related information to enable a “cadas-tral survey” on existing nuclear sites.

A prototype, called SIT (Site Investigation Tool), isunder development with the objective of integrating asmuch as possible site related information from decla-rations and from open sources in a single environment.SIT is an integrated platform for collecting, managingand analysing data from declarations and from a varietyof other sources to support verification of completenessand correctness of states’ declarations. The architec-

20

References

1.11 Mainetti S., Van Paemel F., Cordeil S., Azzalin G.,

Guilmain P., Korn Chr. – Electromagnetic Tests onDr. Neumann’s Digital Camera Module InterfaceDCM-M3 – Technical Note No. I.00/39, JRC Ispra,March 2000.

1.12 Mainetti S., Van Paemel F., Korn Chr. – ElectromagneticTests on Dr. Neumann’s Alis + “Pacpro SurveillanceSystem” – Technical Note No. I.00/63, JRC Ispra, May2000.

1.13 Mainetti S., Van Paemel F., Korn Chr. – “Electro-magnetic Tests on Dr. Neumann’s Portable AlisSurveillance System” – Technical Note No. I.00/138,JRC Ispra, December 2000.


Guilmain P., Korn Chr. – Thermal, Humidity andMechanical Tests on PTS Seal Reader (Standard ClassEquipment – Technical Note No. I.00/10, JRC Ispra,January 2000.


Guilmain P., Korn Chr. – Thermal, Humidity andMechanical Tests on Dr. Neumann’s Alis + PacproSurveillance System (High Class Equipment) –Technical Note No. I.00/40, JRC Ispra, March 2000.

1.16 View of the environmental parameter recording station.

ture of SIT allows the integration of information fromfacility attachments and other data provided in thecontext of the traditional safeguards declarations, aswell as information provided under the new additionalprotocol. Besides verification, SIT may also supportinspection planning and training. A study was startedto investigate the use of SIT in combination with dis-persion modelling to support planning of activities andinterpretation of results related with wide area sampling.

1.7.2 Information Centre on Non-proliferation

L.V. Bril, F. MoustySupport to the definition of a common EU non-prolifera-tion policy as included in the mission statement of theJRC is a very important task. The non-proliferation issuenot only addresses nuclear activities, but also includeschemical and biological material and in general, Weaponsof Mass Destruction (WMD). First, a database is beingestablished to allow for fast answers to requests comingfrom the political world. However, this database cannotbe considered as a simple “bank” because many gov-ernmental organisations (GOs) and non-governmentalorganisations (NGOs) have already information availa-ble on the web. It has been conceived as a dynamicsystem, which analyses and validates information,and it will describe and analyse the present and futurevarious material cycles (nuclear fuel cycle, chemicalindustry etc.). It should, from these investigations,foresee the consequences for Non-Prolifera-tion (NP)and Verification Regimes. Above all, it is conceived asan impartial and independent tool at the service of theEU i.e., independent of external pressure. The tasks ofthe Information Centre can be identified as follows:

• Establishment of a first information set concerningTreaties, Agreements, Regulations and the variousGOs involved in the implementation. A second setwill concern the NGOs. This task is currently welladvanced.

• Establishment of a competence centre on presentand future material cycles (nuclear fuel cycles,chemical and biological industries, technologies,developments, etc) and world-wide monitoring ofactivities (NP, disarmament and WMDs). This task isunder development.

• Establishment of a continuously updated databaseand analysis and validation of the data collected.Part of the information should be available to thepublic through the web or other information means.Furthermore, by using the information on the data-base, country profiles can be established and ana-lysed. This task has begun and some preliminarydocuments have been produced.

• As a function of technological improvements, evalua-tion of the consequences on NPs, verification regimesand disarmament.

1.8 INSPECTOR TRAININGM. Franklin, S. Guardini

Training in nuclear safeguards is expensive for a varietyof reasons. It requires specialised laboratories with anextensive range of instrumentation and nuclear sam-ples. It also involves the time spent by specialists ingiving courses and preparing manuals. For the pastseveral years at ISIS, training and calibration laborato-ries like PERLA and TAME have been created, with theaim of simulating as close as possible field conditions.

ISIS nuclear activities play an important role in provid-ing training in the techniques of auditing of account-ancy for plutonium and uranium in nuclear facilities.This auditing ensures that the nuclear material is prop-erly accounted for. The major client for this trainingservice is ESO. The ISIS training course menu includesabout 15 courses covering, for example, passive andactive neutron measurement methods, gamma spec-trometry techniques for isotopic composition, calorim-etry, mass and volume measurements for liquids in ac-countancy tanks as well as courses dealing with thepractical procedures of verifying an inventory and sub-sequent data processing. In most of the courses forESO, the instruction is a combined effort from ISIS spe-cialists and experienced staff from ESO itself. ESO isnot, however, the only client of these training coursesas some of the ISIS courses for ESO are also attendedby the IAEA inspectors, mainly oriented to solutionmonitoring in the TAME facility. During 2000, ISIS gavea total of ten different types of courses for ESO andIAEA participants.



Poucet A., Contini S. – Use of Geographical InformationSystems (GIS) and Satellite Images for Safeguards –ESARDA Conference, Dresden, May 2000.

1.17 SIT allows the integration of information from declarations

and the comparison of this information with independent

data e.g. maps and satellite images.

A special training course led by personnel from the LosAlamos National Laboratory and supported by ourstaff was held in PERLA. Staff from ESO and the IAEAattended the course. During the course, the attendeesgained an intimate knowledge of the functioning of theAdvanced Multiplicity Shift Register, used for the det-ermination of the mass of fissile material. The coursebenefited from the expertise, infrastructure, and inparticular, the nuclear reference material of the PERLAlaboratory which was used extensively. This equipmentwill be deployed world-wide to assist inspectors intheir day-to-day job of monitoring nuclear material toavoid the proliferation of nuclear weapons.

The emerging technologies of computer based learn-ing can meet some of these needs in an appropriateway. For this reason ISIS has an on-going project of dev-elopment of multimedia modules for ESO inspectors.This development is an active collaboration betweenISIS and the training personnel in ESO and uses ext-ernal software companies as much as possible. Thesemultimedia modules will be made available via network.

The available modules cover active and passive neu-tron measurements, gamma spectrometry, as well asconcepts of measurement uncertainty and measure-ment quality control. The content covers the practicalsetting up of the instrumentation, use of software, theoryof the measurement methods, problem solving, etc.

In the framework of ISIS’s support to the developmentof similar training in Russia, courses are also organisedfor Russian participants who will be instructors in theRussian training centres. During the year 2000, ISISprovided three courses for Russian participants. Thesecovered K-edge/X-ray densitometry of liquid phasenuclear materials, Non-Destructive Assay (NDA), Statis-tics and Nuclear Material Accountancy and Characteri-sation of NDA Reference Materials (together with US-DOE).

1.9 TECHNICAL ASSISTANCE TOTHE COMMONWEALTH OF INDEPENDENT STATES (TACIS)

Co-operation with Russia and Kazakstan is performedin the framework of the various TACIS projects that are co-ordinated by ISIS. The following projects are under way:• Russian Methodology and Training Centre (RMTC

Obninsk) in co-operation with the Institute for Physicsand Power Engineering (IPPE).

• Instrument development (in co-operation with theInstitute for Automation - VNIIA).

• Ural Siberian Methodology and Training Centre(UrSiMTC Snezhinsk) in co-operation with the AllRussian Institute for Technical Physics (VNIITF).

• Development of NMAC (Nuclear Material Accountancyand Control) systems for fuel fabrication plant inUlba (Kazakstan).

• Development of NMAC systems NPP Kursk and Kalinin(in co-operation with Rosenergoatom - REA).

1.9.1 RMTC (Obninsk) and URSIMTC (Snezhinsk)

In recent years one of the interesting spin-offs that hasemerged from the European Commission’s activities innuclear safeguards is the fact that it has become a res-ource for transferring EU experience of nuclear materi-als management to the Russian Federation, which isinvolved in upgrading its own accountancy and controlof nuclear material. Among its support activities underthe TACIS programme, ISIS has assisted the RussianFederation in setting up a training centre at the Insti-tute of Physics and Power Engineering at Obninsk –similar to the PERLA training centre at ISIS. The supporthas included: provision of instrumentation; provisionof nuclear material standards; modification of labora-tories; training; participation in training courses andworkshops at Obninsk and making available ISIS train-ing documentation.

In a second project, work has begun in identifying thedetailed needs for a second training centre in the Uralregion. This centre, UrSiMTC (Ural Siberian Methodol-ogy and Training Centre) will focus on the back end ofthe fuel cycle. A first laboratory to be implemented willaddress the NMAC aspects related to plutonium storage.The detailed specifications for this laboratory have beenworked out.

22


References

1.16 Foggi C. (Ed.) – Proc. of the 22nd Annual ESARDAMeeting – Dresden 9-12 May 2000. EUR Report - tobe published.

1.9.2 Instrument Development

This project is supporting the development of Russianproduced instrumentation for NMAC. The main activityin 2000 involved:• a study to identify needs and priorities for the Rus-

sian’s own production of measurement equipment(instrumentation needs, development priorities, reviewof commercial instruments, QC requirements, etc.),

• specification, development of testing procedures andprocurement of instrumentation to measure uraniumand plutonium.

1.9.3 NMAC at the ULBA facility

(Further details can be found under section 1.1: Mass

and Volume Measurement Techniques.)

The objective of this project “Computerised NMAC sys-tem at Russian nuclear power plants” is to implementa computerised system for NMAC at two Russian nuclearpower plants and then to extend such a system to allRussian nuclear power plants. In 2000, an in-depthanalysis was made of the situation with regard to nuclearmaterial management at the Kursk nuclear power plant(RBMK type) and at the Kalinin nuclear power plant(VVER type).

1.10 ESARDAThe European Safeguards Research and DevelopmentAssociation (ESARDA) remains the most important cor-nerstone of co-operation with other organisations inEurope. In 2000, ESARDA organised its 22nd AnnualMeeting [1.16] in Dresden. This meeting focused on thestrengthening of safeguards and how to integrate tra-ditional verification with the new regime (AdditionalProtocol). The 3rd Joint ESARDA-INMM Workshop washeld in Tokyo. This was the third in a series of work-shops on “Science and Modern Technology for Safe-guards”.

2OVERVIEWI. Shepherd1999 was the first year in which ISIS’s institutionalwork supported EU policies on humanitarian demining.Several projects were started to better understand thenature of the problem and to identify how best the Eu-ropean Commission should focus its efforts in order tobe most effective. In the second year, 2000, many ofthese studies were completed, reports were preparedand delivered to customers. These included a numberof experimental and analytic studies on new detectionmethods – electronic noses, electrostatic and infra-redsensors – as well as algorithms for fusing together in-formation from different sensors and the use of com-puter-aided design packages to distinguish betweendifferent types of mine. A study was completed on thepotential transfer of ISIS’s knowledge on human fac-tors in aviation safety to the training of deminers. Thesecond year also led to a rationalisation of efforts. Thiswas prompted first by insights gained in this work,secondly by requests from the External Relations DG,which co-ordinate the European Commission’s effortson land mines, and thirdly by the arrival in ISIS in October2000 of the “Detection and Positioning Technologies”unit from JRC’s Space Applications Institute (SAI). Theimprovement and assessment of demining technolo-gies became the responsibility of the ex-SAI unit andthe work on information systems began to concentrateon improving the transparency, visibility and effective-ness of EU mine action operations.

The contribution from the “Detection and Positioning

Technologies” unit is described in SAI’s Annual Report

2000 under the section: “Technologies for Detection

and Positioning: Anti-personnel Mines”.

(http://demining.jrc.it/)

2.1 IMPROVING TRANSPARENCY, VISIBILITY AND EFFECTIVENESS OF EU MINE ACTIONSI.Shepherd, R. Peckhamhttp://eu-mine-actions.jrc.cec.eu.int/

In view of the enormous nature of the problem, effort isfocusing on South-East Europe. This has two objectives– first to contribute towards the European Commis-sion’s road map for the region and secondly to developmethodologies and tools that can be applied morewidely. ISIS continued to maintain its web-site and on-line database of EU mine actions. Progressively itsaccuracy and level of detail improved as informationfrom the different services of the Commission involved– the External Relations DG, Common Service for Exter-nal Relations (now AIDCO – EuropeAid Co-OperationOffice), ECHO and the Development DG – were cross-checked and correlated. An on-line geographical infor-mation system was completed.

A major effort was made to harmonise information inthe South-East Europe region. A preliminary workshopin Ispra in March 2000 was followed by another, inSeptember 2000 at the Bosnia-Herzegovina Mine ActionCentre (BHMAC), Sarajevo. Participants included rep-resentatives of Mine Action Centres in Albania, Croatia,Bosnia Herzegovina and Kosovo. Intensive discussionsled to the definition of standard terminology for thecontents and formats of reports describing infectedareas (type, position, estimated area, numbers andtypes of ordnance). The approach used was to keepthe reports as simple as possible to begin with andrestrict these transfers to information which could beplaced in the public domain. The contents and formatsfor reports describing incidents and accidents were alsodiscussed and agreed following similar lines.

A six-monthly reporting interval was agreed. In 2000,the first deliveries of information were made by the MineAction Centres to the JRC. Information was receivedfrom Bosnia-Herzegovina, Croatia and Kosovo. This info-rmation is being fed into a decision support systemlinked to a geographical information system with theprimary aim of helping developers of new technologyunderstand the operational requirements. That thereis a need for such technology is not in doubt. The back-wardness of present equipment – prodders, metal det-ectors, dogs, mechanical means – meant that during2000 only about 10 km2 out of a suspected 4,500 km2

was cleared in Croatia. The decision support systemmerges the mine-field information with informationoriginating in JRC on landcover, slope, population den-sity, climate, soil type, etc. A database is being set upto make this merged information available over theInternet.

The system is intended as a building block for the region-al information system or “headquarters module” wantedby the External Relations DG. This might have securityapplications beyond only mine clearance. In the springof 2000, ISIS participated in a Council of Europe taskforce mission to Kosovo whose objective was to easethe transformation of the former Kosovo LiberationArmy from a military body to a civil protection force.The first step was to prepare an assessment of risks inKosovo – floods, fire, seismic, landmines – that mightneed an intervention from such a body. The JRC contrib-uted its general risk assessment expertise as well as amore specific expertise in landmines and acted as editorsof the final report.

Humanitarian De-mining

24

2.2 REMOTELY CONTROLLED MECHANICALEQUIPMENTF. Littmann

The methods in use today in humanitarian deminingare slow, dangerous and expensive. To develop realistictechniques and procedures that will truly enhance theprocess of mine clearance (faster, safer and cheapermeans), two important factors must be considered:• Mines: the threat presented by the extensive variety

of mines themselves and the ways in which they areused. One of the main threats is from bounding frag-mentation mines and associated tripwires.

• Minefields: the limitations imposed by the real mine-field environment, where the terrain often rules outwhole categories of techniques and equipment. Oneof the main difficulties is the vegetation, which createsa physical access problem (clearing vegetation takesup to 70 % of the deminer’s time), but also makes itdifficult to spot fragmentation mines and tripwires.

Heavy mechanical devices could be the solution, butthey are often derived from military equipment fulfill-ing breaching requirements (high speed, clearancerate around 80% etc), consequently they are heavy(between 10 and 60 t), cause soil erosion, destroy allthe trees or crops and their transport, mobility, spareparts and associated logistics are often not affordablefor most non-governmental organisations. Light mech-anical demining vehicles (between 1 and 5 t) are usedin humanitarian demining, mainly for vegetation cut-ting and tripwire clearance. They are remotely control-led, offer a high mobility, are easy to transport and areaffordable for most operations. A preliminary survey oflight and remotely controlled mechanical equipmenthas provided information on several potential alreadyexisting candidates.

2.3 PILOT CAD DATABASE FOR THE CLOSE-INDETECTION OF LANDMINESF. Andritsos, F. Littmann, I. Vakalis

One of the major issues in landmine detection is thevery high false alarm rate associated with the highsensitivity metal detectors needed for the detection ofthe modern minimum metal mines, coupled with theusually high debris soil contamination. At present,each metal detector alarm can only be confirmed bymanual probing and digging to visually identify thesuspect object.

The work performed at ISIS aims at using 3D ComputerAided Design (CAD) models of mines for a pilot multi-sensor data-fusion application. CAD feature extractioncoupled with multi-sensor data fusion algorithms isexpected to help in identifying the cause of many metaldetector alarms, thus saving a lot of manual work andenhancing the demining process significantly.

Continuing the work started in 1999, staff memberswere trained to use the CAD software tool, which hadbeen chosen as being the most suitable for the envis-aged work and several CAD models of anti-personneland antitank mines were done.

2.4 HUMAN FACTORS IN HUMANITARIAN DE-MININGP.C. Cacciabue

Despite the amount of research conducted in order toidentify new technologies to support clearance opera-tions in humanitarian de-mining, detection and clear-ance are still performed manually using metal detectorsand probes. The human being is the main actor in fieldoperations.

A pilot research started in early 2000 helped in identi-fying performance indicators and professional compe-tence of deminers [2.1]. These results, in turn, definethe training needs in terms of objectives and contentsto be developed and emphasised in a training courseon non-technical issues, i.e., human factors. The mostimportant findings of this pilot research can be sum-marised in the definition of two levels of interventions:risk management and individual-organisational fac-tors (Figure 2.3). At risk management level, the mainfactors which have an impact on de-miners’ behaviour


2.1 A few examples of remotely controlled light mechanical

equipment.

2.2 PMA2 CAD model and real anti-personnel mine.

are related to attention, risk perception, national/safetyculture, mental and physical fatigue (normal durationof a single prodding task should not exceed 20 minutes).In particular:• The level of attention of de-miners decreases strongly

with routine and complacency and this causes “poor”or unsafe results.

• Risk perception, decreased by routine as well as thehigh rate of false alarms (100-1000 false alarms foreach real mine detected), further increases “over-confidence bias” in detection and inspection tasks,as well as “invulnerable” and “macho” attitudes.

• National/safety culture determines the way in whichStandard Operational Procedures and safety equip-ment are applied and used. For example, in somecultures, wearing protective equipment is seen as asign of weakness.

• Fatigue can strongly affect performance during prod-ding and de-miners can normally carry out this taskfor 20 minutes before requiring a rest.

The most important individual and organisationalfactors that may be the objective of training are:• Teamwork, highlighting the role of good communi-

cation among team members, team building, andinterdependence.

• The features of successful leadership.• National culture as way of communication and rela-

tionship with the local population.• Co-ordination and co-operation within the organisa-

tion and among the different institutions dealing withdemining in the same area.

In the future, the research work will focus on the long-term objective of developing guidelines for the design(objectives, contents, and methodology) of human fac-tors training for personnel working in the field ofhumanitarian de-mining.

In the framework of humanitarian de-mining, a collab-oration with the University of Padua and the Universityof Turin focused on the study of professional compe-tence of experts in manual clearance operations. Inparticular, the Naturalistic Decision Making (NDM)movement, initiated in the United States in 1989, ad-dresses the study of how people use their experienceto take decisions in the field setting. NDM researchhas been mainly applied in two main domains: pilots(aviation) and fire fighting brigades. The collaborationwith the Department of General Psychology of the Uni-versity of Padua showed that the NDM paradigm andmethodologies can be applied for the evaluation of themetal processes of typical experts in the humanitariandemining domain. Consequently it can be exploited forthe selection processes and for the development oftraining [2.2].

A pilot research was planned and carried out in collab-oration with the University of Turin – Department ofPsychology – and the Italian Non-Government Organi-sation (NGO) Intersos. The pilot research aimed atidentifying elements of professional competence ofexperts working in the demining field. Different method-ologies were taken into consideration and some wereimmediately rejected, for instance observation and“think aloud”, due to the specific type of competenceand the difficulty in their application in the field duringclearance operations. Finally, three different tech-niques based on verbal interactions were selected andapplied: instructions for one’s double, behaviouralevent interview and story telling. According to the pre-liminary results obtained, the research highlighted somemajor areas of professional competencies, namely: so-cial/interpersonal and cultural dimension (relation-ship with the team-members, the local population andlocal organisations/institutions); and cognitive dimen-sion (in particular, decision making competencies) [2.3-2.4]. These data will guide the second part of the re-search, which will be conducted in collaboration withthe Italian Army involved in humanitarian de-mining.The final results will lead to the identification of theelements of professional competence, which shouldbe trained in order to increase efficiency and safety ofde-mining operations.

References

2.1 Bacchi M., Cacciabue P. C., Bellezza M., Re A. – HumanFactors and Professional Competence in HumanitarianDe-mining – 10th European Conference on CognitiveErgonomics, ECCE-10. Linköping, Sweden, August21-23, 2000.

2.2 Bastianini, G. – Humanitarian De-mining: a naturalisticapproach to decision making – Graduation Thesis in Psy-chology (in Italian), Università di Padova, Italy (2000).

2.3 Bellezza, M. – Humanitarian De-mining: from techno-logical improvement to valorisation of deminers’professional competence – Graduation Thesis in Psy-chology (in Italian), Università di Torino, Italy (2000).

2.4 Donato, E. – Contributing to the qualification of de-miners by an investigation on the development ofcompetence – Graduation Thesis in Psychology (inItalian), Università di Torino, Italy (2000).

2.3 Human factors training issues for humanitarian demining

National SafetyCulture

Mental and PhysicalFatigue

Routine andComplacency

Risk Perception

Co-operation

National Culture

Teamwork

Leadership

Human-centered approach is necessary to study de-miningactivity, coupled with an ergonomic analysis

Risk Management - Trainnig Issues

26

3 Dependability of Information Technology Systems 28

3.1 Creating Consumer Confidence in e-commerce 283.2 On-line Privacy 283.3 Children and the Internet 283.4 Security of Critical Information Infrastructures 293.5 DRIVE (Drug in Virtual Enterprise) 293.6 TRINIDAD Test Laboratory for Information Security 29

4. Networks, Multimedia and Education 30

4.1 Web Technologies 304.1.1 European Technology Transfer Network 304.1.2 Information Highways and New Technologies 304.1.3 Virtual European Statistical Laboratory 314.1.4 European Treasury Browser 31

4.1.5 ParlEuNet – European Parliament Network 324.1.6 Solar Radiation Databases 324.1.7 Support to Improving Human Potential Programme 33

4.2 Support for the Disabled 334.2.1 Rehabilitation IT Aid for the Parkinsonians (PARREHA) 334.2.2 Adaptive Brain Interfaces 334.2.3 SMART-2 344.2.4 VOICE: Giving a Voice to the Deaf 354.2.5 ACCESS to Education for People with Disabilities 364.2.6 Multimedia Simulation System for the Disabled 364.2.7 Improving of Error Diagnostic Skills in the Disabled 37

5. Medical Imaging – Data Systems 38

5.1 Moving from 2D Data to Real 3D Medical Imaging 385.2 Scientific and Technical Support for Medical Images and Systems 39

6. Applied Statistics 40

6.1 Network Building in Collaboration with EUROSTAT 406.2 Convergence of EMU Economies 406.3 Short-term Economic Indicators 406.4 Economic Time Series relevant to EMU and EU Areas 416.5 Analysis of Financial Risk 416.6 Environmental Models and Indicators 416.7 Dissemination 42

7. Knowledge Assessment Methodologies 44

7.1 Science and Governance 447.2 Quality Assurance of Policy-Related Research 447.3 Information Technologies and Knowledge Assessment 447.4 VISIONS - Integrated Visions for a Sustainable Europe 457.5 GAS - Greenhouse Gases Meter, Scenario Explorer and Day Planner 457.6 GOUVERNe 45

8. Fight against Fraud 46

8.1 Anti-fraud Information Management 468.1.1 Early Warning Analysis 468.1.2 Name Data Cleaning and Information Portals 468.1.3 Open Source Intelligence 478.1.4 Language Engineering 478.1.5 Dissemination 48

8.2 Data Analysis and Risk Analysis for Anti-fraud 488.3 Advanced Statistics for the Clearance of Accounts 498.4 IDEA (Identification Electronique des Animaux) 508.5 New Technologies for Monitoring Fishing Vessels 52

CYBER-SECURITY, KNOWLEDGE ACCESS ANDASSESSMENT, THE FIGHT AGAINST FRAUD

27Cyber-security, Knowledge Access & Assessment, the Fight against Fraud

OVERVIEWM. WilikensThe explosive worldwide growth of the Internet, itsvulnerabilities and the lack of clear legal rules in inter-national electronic commerce (e-commerce) have raiseda legitimate concern with respect to the adequacy ofsecurity and consumer protection measures. The researchduring 2000 addressed key concerns in the areas ofconsumer trust and confidence in cross-border elec-tronic commerce, safeguarding privacy in the on-lineenvironment and protecting critical information infra-structures from vulnerabilities and cybercrime. ISIS hasso far concentrated its support activities in the cyber-security area with the Information Society DG. However,the impact of electronic technologies is so great that awide range of policy-making DGs also needs experttechnical advice in defining their policies and recom-mendations. These include the Health and ConsumerProtection DG, Internal Market DG and the EnterpriseDG. Similarly the European Parliament has shown inter-est in impartial advice regarding technical constraintsand implications of proposed legislation. In the EuropeAction Plan endorsed at the Feira European Councilmeeting in June 2000, the EU underlined the impor-tance of building up consumer confidence on the Internet(eConfidence) in order to accelerate growth in businesstowards consumer e-commerce.

3.1 CREATING CONSUMER CONFIDENCE INE-COMMERCE BY MEANS OF EFFECTIVEREDRESS MECHANISMSM. Wilikenshttp://econfidence.jrc.it/

Whilst e-commerce has many qualities that consumersfind attractive, it also has properties that facilitate fraudand make prosecution in the judicial system difficult.Business to consumer e-commerce is characterised by alarge volume of relatively low value transactions carriedout across borders. Thus, the relevance and effectivenessof courts for resolving problems may be limited. Alterna-tive dispute settlement mechanisms for providing con-sumer redress are therefore needed. New web-basedtechnologies offer new possibilities, features and dimen-sions to dispute settlement. A successful workshop inMarch 2000 (in collaboration with other CommissionDGs, industry and consumer organisations) exploredtechnical concepts and criteria for on-line dispute settle-ment systems that are fair, effective and secure for resolv-ing cross-border consumer disputes at a pan-Europeanlevel. As a direct consequence of the workshop, an on-lineeConfidence forum (http://econfidence.jrc.it/) was devel-oped to facilitate information exchange and discussionson eConfidence. A new project to develop a demonstratorplatform for testing components, business models andsoftware applied to on-line consumer dispute settlementwas launched later during 2000.

3.2 ON-LINE PRIVACYT. Jacksonhttp://dsa-isis.jrc.it/Privacy/

The Internet and the World Wide Web have made itpossible for an individual's activities to be monitored,tracked and profiled closer than ever before. In Europe,efforts have been made to create a legal framework toprotect the privacy of personal data collected or storedon-line. Nevertheless, legislation can only play a partialrole in securing data privacy. To overcome this, ISIS isconducting a project for the Information Society DG onhow technology can support European data protectiondirectives. The project is investigating the state-of-the-art in privacy technologies for consumer and corporatedeployment, and identifying privacy threats inherentin new and emerging technologies such as mobile code(software agents), biometrics, e-money and mobiletelecommunications. In support of these activities astudy and a workshop entitled “The Role of Technologyin Facilitating On-line Privacy” was held in Brusselsduring May 2000. Consequently, a Privacy Hub (http:

//dsa-isis.jrc.it/Privacy/) was set up. Reference imple-mentations of the Platform for Privacy Preferences (P3P)standard for privacy management will be tested in theTRINIDAD laboratory.

3.3 CHILDREN AND THE INTERNETT. Jacksonhttp://efilter.jrc.it/

Effective filtering techniques and web tools for use byparents and teachers are a means to allow controlledaccess to high quality educational material and toavoid access to defamatory content. Nevertheless, theperformance of these tools, both in terms of usabilityand effectiveness, is variable, and often does not meetthe requirements of the individual users. ISIS has com-menced a study, including laboratory tests, to foster abetter understanding of the process of benchmarkingfiltering software and services used to protect childrenfrom harmful or undesirable content on the Internet.In particular the study seeks to define the evaluation

28

Dependabilityof Information Technology Systems3

3.1 Prototype of consumer dispute handling system.

criteria that might be employed in specific benchmark-ing exercises. The project is intended to further theactivities of the European Commission’s Safer InternetAction Plan, in particular action line 2 (developingfiltering and rating systems) and to contribute to thedefinition of the EU research agenda in this area.

3.4 SECURITY OF CRITICAL INFORMATIONINFRASTRUCTURESM. Maserahttp://deppy.jrc.it/

Society is increasingly dependent on the proper func-tioning of critical large-scale information infrastructuresthat underpin e-commerce, education, health care, etc.These infrastructures are vulnerable to malfunctions,attacks and cybercrime and are very complex due toextensive interdependencies. Initial technical discuss-ions on possible transatlantic collaborations on the issuewere organised by ISIS for the Information Society DGin 1999. Subsequently, an exploratory study was per-formed on “Dependability and Complexity: Exploringideas for studying large unbounded systems”. Anumber of industrial sectors representing utilities, tele-communications and health care contributed to thiswork. An electronic forum was developed to fosterinformation exchange and networked collaboration onthis important societal issue.

3.5 DRIVE (DRUG IN VIRTUAL ENTERPRISE)M. Wilikenshttp://dsa-isis.jrc.it/

The Drive project is a shared-cost action research projectunder the Information Society Technologies programme,which commenced in July 2000. It seeks to develop asecure information-sharing infrastructure to improvethe exchange of drug-related information between phar-maceutical companies and the various departments inhospitals concerned with therapy administration topatients. The objective is to increase patient safety byreducing drug-related errors by means of appropriateintegrity checking of the information and identificationof the people involved, whilst protecting privacy con-temporaneously (Figure 3.2).

3.6 TRINIDAD TEST LABORATORY FORINFORMATION SECURITYL. Breitenbach

ISIS is further developing the TRINIDAD (Trial Infra-structure for Dependable Applications Deployment)laboratory for testing and assessing information securitysolutions that increase business and consumer trust ininformation society applications. TRINIDAD is an openinfrastructure based on open computing and commu-nications standards (TCP/IP, WWW, Java, Open Source,etc) to ensure platform independence of the testedsolutions. The tested solutions include newly developedtechnologies, best practices as well as reference im-plementations of emerging standards (e.g., P3P stand-ard for privacy management). Demonstrators have beendeveloped on TRINIDAD using European Java-basedtechnology. One demonstrator concerns a kiosk for thesecure retrieval by patients of medical documents.Another is a Java-based smart card test platform forweb-centric applications. The platform will amongstothers be used for evaluating electronic authentica-tion and identification techniques in the DRIVE project(section 3.5). Furthermore, TRINIDAD is being equippedto perform benchmarks of web filtering tools to protectagainst harmful Internet content (section 3.3).

Cyber-security, Knowledge Access & Assessment, the Fight against Fraud 29

3.2 Secure healthcare IT infrastructure for patient safety (reduce

adverse drug events) whilst protecting patient privacy.


Kyriakopoulos N., Wilikens M. – Dependability of com-plex open systems: A unifying concept for undersanding Internet-related issues – Third InformationSurvivability Workshop, Boston, Massachusetts,USA, 24-26 October 2000.

Jackson T. – The role of technology in facilitatingon-line privacy - EUR 19643 EN (2000).

Wilikens M., Vahrenwald A., Morris P. – Out-of-courtdispute settlement systems for e-commerce – EUR19644 EN (2000).

Wilikens M., Jackson T., Sanna A. – Dependability re-quirements of large-scale information infrastructures:A case study from the health care sector – EUR 19642EN (2000).

Breitenbach L., Wilikens M. – Smart Card Test Infrastruc-ture for Web-centric Applications – Second South-Eastern Conference on eCommerce, Sofia, Bulgaria,24 - 26 October 2000.

Jackson T. – Benchmarking of Filtering Software andServices: An Analysis Framework – IST conference,Nice, France, 6-8 November 2000.

Jones S., Wilikens M., Morris P., Masera M. – Trustrequirements in e-Business – Communications of theACM (Association for Computing), Vol. 43, No. 12,December 2000.

3.3 TRINIDAD Smart card application in the health care sector.

Web ServerClinical

ApplicationServer

Database

SecureEnvironment

Web BrowserDoctor

SSL-ClientCertificateretrievedfrom Card

BrowserPlug in

Hidden JavaWorkerApplet

Card Readerwith Card

SmartCard

SecureEnvironment

TCP/IP

TCP/IP

SSL

Secure Channel

30

4.1 WEB TECHNOLOGIESC. Best

An important impact of the web on society is the newpossibility for large-scale distance education and train-ing. In a fast changing world, the need for lifelong learn-ing, the development of educational technologies andmeasures to combat social exclusion are all recognisedas major EU policies. During the year 2000, the researchcarried out in ISIS supported EU policies in the area ofweb-based education and training in the following mainareas:• Technical co-ordination of the PROmoting Multimedia

access to Education and Training in EUropean Society(PROMETEUS) network – a European Commissioninitiative encouraging collaboration and standardisa-tion between over 400 companies and organisationsin developing educational multimedia for Europe(http://prometeus.org/).Development and prototyping of the Gateway toEuropean Learning Area.

• Technology observatory: To provide technical adviceto the Information Society DG in developing theirpolicies in education and training through hands onprototyping.

The main achievements were:• Successful running and maintenance of the PROME-

TEUS web service and 12 mailing lists used by thespecial interest groups (SIGs).

• Generic Information Server Toolkit (GIST) softwarereleases V1.04, 1.05 and 1.06.

• Gateway demonstrator (http://learningeurope.org/)

was developed and demonstrated at an intergovern-mental conference in Lisbon and at the Hannovertrade fair.

• Knowledge Management System was developed asa simple fast but powerful community forum andweb-site. It was demonstrated to an inter-DG work-shop on Knowledge Management (figure 4.1).

• Development of a virtual live meeting system – Power-comm – and an operational web-casting system,which were demonstrated to the Education and Cul-ture DG and the Information Society DG at a jointworkshop.

4.1.1 European Technology Transfer Network (ETTN)

C. Segrehttp://ettn.jrc.it/

ETTN was a European Commission initiative designedto make technical research and expertise widely avail-able to European industry. The aim of ETTN was to impr-ove the process of technology transfer by encouragingthe exchange of information and the setting up of aneffective network of services. The technical manage-ment of the project was performed by ISIS on behalf ofthe Information Society DG and the Enterprise DG andwas completed in September 2000.

ETTN facilitated technology transfer among Small Me-dium Enterprises (SMEs) belonging to informationtechnology and environment sectors using a demand-oriented approach supported by a wide use of web-based communication tools. More than 370 SMEsbelonging to 14 clusters, and technology brokers withan extensive knowledge of local industry tried tomatch technology requests and technology proposalsexchanged throughout Europe using a web-basedcommunication system. This new web-based system(a combination of GIST and CIRCA software) was spec-ifically developed for the project. At the end of theproject, 233 technology requests and 171 technologyproposals had been introduced in the system. 27 ETTNcase studies (i.e., actual contractual agreements be-tween trans-national partners) have been published.Many organisations, announcements, innovations,events, news, services and products dealing with SMEsand technology transfer have been introduced in theETTN web-site to facilitate the information exchangebetween SMEs.

4.1.2 Information Highways and New Technologies

P. Loekkemyhrhttp://vesl.jrc.it/

EUROSTAT recognised early on that the rapidly grow-ing Internet and the wealth of new and emerging tech-nologies associated with it represented an unprecedent-ed opportunity for improving access to, interchangeand visualisation of statistics related information. Thisproject aimed at reviewing some of these new andemerging technologies, selecting promising contendersand developing and evaluating possible implementa-tions in the form of demonstrator systems. ISIS has, incollaboration with other institutes in the JRC, namely,the Environment Institute (EI) and the Space Applica-tions Institute (SAI), investigated a range of technolo-

Networks, Multimedia and Education4

4.1 The Knowledge Management GIST theme used

for e-Commerce.


gies: desktop video-conferencing; streaming video onthe Internet (figure 4.2); tools for virtual meetings; inter-active visualisation of urban land use in virtual reality(SAI); visualisation of geo-referenced data (EI); andweb-based Geographical Information Systems (EI).Prototype systems have been developed to identifyuseful features and possible applications. Resultsfrom surveys and experiments have been used to pro-duce guidelines for efficient implementation of infor-mation systems employing these technologies. The res-ults and technologies from the project will be fed intothe technology transfer activities that ISIS runs in col-laboration with EUROSTAT (see section 6.1).

4.1.3 Virtual European Statistical Laboratory

P. Loekkemyhrhttp://time-series.jrc.it/

The Virtual European Statistical Laboratory (VESL) is aproject undertaken by ISIS for EUROSTAT. The goal forthe first half of the project was to design and implementa community-style Internet service for research andtraining in official statistics. The service was put intooperation in April 2000. The service contains directo-ries of people and organisations, libraries for softwareand documents and interactive services, which enableusers to announce events and share software, dataand documents with the community. Maintenance ofthe service is mainly done in a decentralised mannerby the users themselves; VESL is an open self-populat-ing web-site. The service is built using Generic Infor-mation Server Toolkit (GIST) technology (figure 4.3).

Towards the end of 2000, the project continued to pursueobjectives related to demonstration, dissemination andtechnology transfer by broadening the scope to includesome results from the Information Highways and NewTechnologies project (section 4.1.2). The project willcontinue until the middle of 2001 and the project results

will be fed into the technology transfer activities thatISIS runs in collaboration with EUROSTAT (see section6.1).

4.1.4 European Treasury Browser (ETB)

A. Reggiorihttp://etb.jrc.it/

http://rdfstore.jrc.it/

http://mda07.jrc.it/

http://xml.jrc.it:8080/

The European Treasury Browser (ETB) is the EuropeanSchoolnet’s web-based educational resource infrastruc-ture for schools in Europe and will provide a European-wide clearing house for educational resources. Theproject started in February 2000. ETB is based on a dis-tributed metadata network connecting repositories ofeducational web resources across Europe. The ETB archi-tecture is designed to provide high quality informationsubmitted from each agency. To support publishing andaccess to high-quality educational resources, the en-visaged components for ETB are as follows:• A web-enabled multilingual educational subject

classification and thesaurus to aid accessing andproviding content.

• An intelligent data-entry system for the end-user in-cluding a metadata authoring tool with gateways toexisting metadata systems and a quality assuranceprocedure.

• A dynamic metadata network to allow the flow ofinformation across the Internet.

• A metadata registry with an intuitive search interface.• A full set of measures of harmonisation and normali-

sation addressing different layers.

4.2 Streaming video on the Internet was one of the topics of

the project. The final presentation, live video and presen-

tation screen of the project was webcast live in November

2000 to ten viewers in different European countries.

4.3 Example page from the web-service used to support

research and training in the area of time-series in statistics.

4.4 ETB system architecture – an educational metadata

network for European schools.

NNTP

Manager

Thesaurus

ETB NativeRepository

ETB Registry

Site CSite B

Site A

Site E

Site DNational

RepositoryETB/LocalMapping

ETBInterface

32

4.1.5 ParlEuNet – European Parliament Network:

a student’s parliament via educational

multimedia learning

A. Reggiorihttp://pen.jrc.it

http://parleunet.jrc.it

The ParlEuNet project began in January 1998 and offi-cially terminated in June 2000. The ParlEuNet projectwas the first European initiative to permit secondaryschool students to use state-of-the-art networks andmultimedia resources to learn about and undertakecollaborative projects on the European Parliament. Inter-net connections, videoconferencing and a web-spacecontaining a well-structured updateable multimediadatabase of educationally relevant materials was usedby students to access information on the Parliament,create their own projects, and exchange informationand views with members of Parliament and students inother countries.

The ParlEuNet system provided a multimedia databaseon the Internet, which was accessed by 12 schoolsfrom 7 European countries. The theme of the databasehas been the history, institutions and functions of theEuropean Parliament and the “The world of work”. Thesystem can handle two types of information: referencematerial and dynamic educational material. ISIS devel-oped a highly fast and efficient dynamic database tostore, browse and retrieve such multimedia information.Cutting-edge technologies were deployed and investi-gated to actually implement the final product: XMLdata modelling; custom built, fast DBMS server usingBerkeleyDB files; HTTP state-fullness and caching; liveHTML document editing; user profile and individualsession management; Apache, mod_perl and Msql;and UNIX FreeBSD 100Mbit interconnecting fabric.

4.1.6 Solar Radiation Databases

M. Kleih, C. Besthttp://soda.jrc.it/

http://opengis.jrc.it/

The project Solar Radiation Databases (SoDa) answersthe needs of industry and research for information onsolar radiation data. ISIS has developed a prototypeJAVA tool for creating maps from multiple web servers.This is the first stage in creating an “Intelligent System”that should be able to process user from professionaland non-professional solar radiation data users overthe Internet. The ISIS JAVA tool implements the Inter-operable Web Mapping Testbed (WMT) standard fromthe OpenGIS Consortium (http://opengis.jrc.it/). Thisstandard defines how map servers respond to requests.The data can be distributed over the Internet and isaccessed through HTTP requests (URLs.). For orderingthe geographical databases, another protocol devel-oped by ISIS, the “HTTP-based geotemporal Search”(HGS) was used. This protocol allows one to define ahierarchical structuring of databases whereby the def-inition of the structure can be distributed. It also ena-bles the participating institutes to maintain their ownbranch in the database hierarchy. To simplify partici-pating data to the Geographic Map Request Tool (GMRT),another system was developed to upload and/or regis-ter geographic data. (http://soda.jrc.it/) All registereddatabases appear automatically as WMT servers. In thefuture a distributed processing system will be incorpo-rated, which will permit users to run algorithms to furtherprocess solar radiation data.

4.5 The web desktop developed for ParlEuNet.

4.6 The web-mapping software developed for the

SoDa project.


4.1.7 Support to Improving Human Potential

Programme

C. Best, P. Henshaw, P. Shielshttp://improving.jrc.it/

http://locin.jrc.it/

ISIS has implemented the main web database systemsfor the Improving Human Potential Programme in sup-port of the Research DG. These dynamic web databasesallow users to search for opportunities, networks, vac-ancies, conferences etc. and for remote managers toadd and modify the content all through a web interface.The main services provided are as follows: researchnetworks and vacancies; targeted socio-economic re-search; Marie Curie fellowships; access to researchinfrastructure; and high level scientific conferences. Alarge new system was released in December 2000 topresent the database on “Local Initiatives to CombatSocial Exclusion” (LOCIN). All these systems have beenbuilt using the JRC’s Generic Information Server Toolkit(GIST) technology, which has a proved robust and reli-able framework for this type of dynamic website andoffers users sophisticated search interfaces.

4.2 SUPPORT FOR THE DISABLED

4.2.1 Rehabilitation IT Aid for the Parkinsonians

(PARREHA)

F. Andritsos, I. Vakalishttp://www.parreha.com/

Parkinson’s disease is a dysfunction at the level of motorplanning. Recently, scientists have shown that visualstimulation of Parkison’s disease patients with “virtual”obstacles and auditory stimulation with rhythmicalsounds can significantly improve the efficacy of tradi-tional rehabilitation schemes. The project aims at thedesign and development of a system for:

• the assessment of motor performance,• design and performance of training programmes

for patients suffering from Parkinson’s disease byincorporating virtual reality (VR) visual stimulation,auditory biofeedback and interactive video conferencetechnologies to conventional kinematics analysis andrehabilitation exercise protocols.

The project contributes to social policies by guaran-teeing a high level of social security related servicesthroughout the community. Such activities stronglymotivate human mobility, which is endorsed by allEuropean policies. ISIS’s role in the project is that of amultidisciplinary research centre, which, independentof commercial or other interests, contributes to the userrequirements; lays down the functional requirementsof the system as well as the acceptance criteria basedon the user requirements; presides over the validationcommittee (medical and VR experts and Parkinson’sdisease patients external to the project); and preparesfor an eventual certification procedure. A rapid proto-type of the PARREHA VR training tool has shown impres-sive results in bringing a person suffering from Parkin-son’s disease from the “off-state” to the “on-state”.

4.2.2 Adaptive Brain Interfaces

J. del R. Millán, J. Mouriño, M. Franzéhttp://sta.jrc.it/abi/

The Adaptive Brain Interfaces (ABI) project has devel-oped a portable brain-computer interface based on theanalysis of electroencephalogram (EEG) signals. ABImakes it possible for people to interact with computer-based systems through conscious control of theirthoughts after a short training period [4.1-4.2]. Theportable brain interface has an embedded neural net-work that recognises which mental task the wearer isconcentrating on. It does so by analysing continuousvariations of EEG rhythms over several cortical areas ofthe brain. The approach is based on a mutual learningprocess where the user and the ABI interface are cou-pled together and adapt to each other. Interaction isquite natural as the user can make spontaneous andself-paced decisions without having to wait for, or res-pond to, external cues. The user can also concentrateon a wide range of mental tasks, from motor-related(e.g., imagination of physical movements) to cognitivetasks (e.g., completing mental operations involvingawareness and judgement). Analysis of learned EEGpatterns confirms that for a subject to operate satis-factorily his/her personal ABI, the latter must fit theindividual features of the former. Briefly, there arehardly any features shared by subjects [4.3]. Buildingindividual interfaces greatly increases the likelihoodof success, as demonstrated for all subjects we haveworked with despite the short training time of most ofthem.

4.7 Entry Page to the LOCIN Database developed by ISIS.

During 2000, several applications were developed thatillustrate the wide range of systems that ABI can belinked to. Thus, ABI augments disabled people’s com-munication capabilities through the use of a virtualkeyboard, provides new forms of education and enter-tainment (Figure 4.8), and also enables the operationof physical devices such as wheelchairs (see section

4.2.3). The brain-operated virtual keyboard was pre-sented publicly at a workshop on “Biosignal-basedHuman-Computer Interaction” held in Helsinki on 16 May2000. During a live demonstration, a member of our teamwrote a sentence correctly (suggested by the public) ina few minutes. This was recorded by Finnish televisionand shown on the news that evening (Figure 4.9).

The virtual keyboard and the computer game were alsoshown during the IST 2000 Exhibition in Nice on 6-8November 2000 (this is the European conference ofthe information society community organised by theInformation Society DG). In addition to live demoscarried out by members of the project, several visitorstried ABI and achieved rather good performances inless than 1 hour of training. In December 2000, the BBC(British Broadcasting Corporation) invited our team toLondon to work with a disabled person who had volun-teered to validate ABI as part of a service about ourproject. After a few hours of training, he mastered twotasks very well. At the end of the session, the journalistsasked him for his opinion, and what he said rewardedthe team more than anything else: “this is the firsttechnology I have tried (including voice recognition)that made me feel independent and empowered”.

4.2.3 SMART-2

J. del R. Millán, L. Bort, S. ChiappaThis research network aims at training promisingyoung researchers in the broad field of robotics tech-nologies. ISIS’s participation focused on two topics:• the use of ISIS’s Adaptive Brain Interface (ABI) to

control a mobile robot: user’s mental states are as-sociated to high-level commands autonomously ex-ecuted by the robot. The robot relies on reactivestrategies to move in the desired direction in a safe(avoiding collisions) and smooth way.

• Application of mathematical tools to analyse ABIwhen used for human-robot interaction. The analysisprovides a more solid formulation and understandingof the learning paradigm, thus improving ABI’s per-formance and reliability.

This work demonstrates that it is possible for a physi-cally disabled person to operate a motorised wheelchairwithout external assistance provided the wheelchairhas some on-board sensory capabilities. Figure 4.10

shows how ABI enables the operation of physicaldevices, such as a mobile robot that closely mimics the

34

References

4.1 Millán J. del R., Mouriño J., Babiloni F., Cincotti F.,

Varsta M., Heikkonen J. – Local neural classifier forEEG-based recognition of mental tasks – IEEE-INNS-ENNS International Joint Conference on NeuralNetworks (2000). Outstanding contribution.

4.2 Varsta M., Heikkonen J., Millán J. del R., Mouriño J.

– Evaluating the performance of three feature setsfor brain-computer interfaces with an early stoppingMLP committee – 15th International Conference onPattern Recognition (2000).

4.3 Millán J. del R., Mouriño J., Cincotti F., Varsta M., Heik-

konen J., Topani F., Marciani M.G., Kaski K., Babiloni F.

– EEG patterns associated to spontaneous executionof mental tasks. NeuroImage – 11(5): S78 (2000).

4.9 Opening of the service on ABI broadcast by Finnish televi-

sion. The brain-operated virtual keyboard can be used to

select letters from a virtual keyboard on a computer

screen and write a message. As the user concentrates on

different mental tasks, the keyboard is successively split

in smaller parts until a letter is selected.

4.8 Brain-operated computer game. Here the user interacts

with a computer game, the classical Pacman, using only

two commands to make it turn left or right with respect to

the current direction of movement. Otherwise, Pacman

moves forward until it reaches a wall where it stops.


Miesenberger K. – The VOICE Project – Proceedings ofICCHP2000 International Conference on ComputersHelping People with Special Needs, Universität Karl-sruhe (TH) Germany, 2000.

Mezzanotte A. – The Deaf and Television, Deaf Users’Needs for Television Subtitling – Report on the Web ofthe Conference Handicap and Communication withRAI and the European Broadcasters and on related ac-tivities carried out by CECOEV in year 2000.

Paulet J., Pirelli G. – Le Sous-titrage adapté in Canada –Report on the Web of the meetings in Montreal in July2000.

Paglino A., Pirelli G. – The user needs in conferences sub-titling – Report on the Web, 2000.


operation of a wheelchair. In this case, the user giveshigh-level commands (i.e., move forward, stop, turn left,turn right) and all the low-level details are handledseparately. All high-level commands are sent to a reac-tive controller for execution. In this kind of controller,on-board sensors are constantly read and determinethe next action, i.e., change the speed of the wheelmotors.

4.2.4 VOICE: Giving a Voice to the Deaf

G. Pirellihttp://voice.jrc.it/

A significant number of Europeans suffer from hearingdifficulties. They gain no benefit from telephone, non-subtitled television broadcasts or oral presentations.ISIS is encouraging the development of new equipmentand attempting to raise awareness among the hearingpopulation.

The project VOICE demonstrates how speech recogni-tion systems in conversation, conferences and schoollessons can translate voice into PC screen messages,taking into account the requirements of citizens withhearing impairments. In order to develop awarenessamong users, systems producers and services provid-ers, ISIS has organised and presented VOICE at a largenumber of international workshops and congresses toapproximately 5,000 participants, from industry, aca-demia, broadcasting organisations, as well as final users.The prototype demonstrator has been used in severallanguages to generate live subtitling of these events,as an important means of validation in the field. A work-shop has demonstrated the possibilities of the systemto RAI (Radio Televisione Italiana) and other televisionbroadcasters considering the best use of this technol-ogy. RAI broadcast information on the event and onthe prototype’s tests in the schools. The users feel en-couraged and helped by the VOICE team in their con-tacts with the producers and the service providers. For

|

instance, RAI started subtitling the Evening News asan answer to impaired users’ requests (supported alsoby means of this project). ISIS has performed an over-view of television subtitling in Europe and extended itto Canada. It has also investigated the use of this tech-nology in subtitling telephone communication and hasheld a few meetings using subtitled video-conferencesections.

VOICE was initiated as an ISIS exploratory researchproject and then sponsored by the Information SocietyDG’s Telematics Applications Programme (TAP). Theproject involves the active participation of severalassociations of people with hearing impairments, fromdifferent European countries, exchanging information viaa multilingual VOICE Web Forum at http://voice.jrc.it/

for spreading results and best practices. The presenta-tion follows the current recommendations for creatingInternet contents that are accessible by people withvisual impairment or reduced mobility, as an exampleof a web-site designed following the rules of “web-accessibility for all”. A hearing impaired auxiliary staffand a visually impaired student contributed to theweb-site’s development.

4.10 Brain-actuated control of a mobile robot that closely mimics the operation of a motorised wheelchair.

4.2.5 ACCESS to Education for People with Disabilities

G. Pirellihttp://voice.jrc.it/access/

http://voice.jrc.it/moise/

The project ACCESS investigates the development ofmultimedia applications for helping people with disa-bilities in accessing the Information Society in theireducation and training. The aim is that of improvingthe quality of life by trying to include a portion of thepopulation that is currently, at least partially, excluded.The research will facilitate the integration of peoplewith special needs, including the elderly and immi-grants. Owing to a “design for all” approach, everyonewill benefit from this advance.

ACCESS is extending the VOICE project’s basic ideathat the Internet is also the written voice of a teacheror a friend for those who cannot hear the spoken voice.The idea is even larger, in the sense, that also people

with visual impairment may read the spoken web pagesvia text-to-voice synthesis of the written web pages.Workshops, videoconferences, tele-education ses-sions, Net-meetings and an ACCESS Web Forum offeropportunities for sharing common experiences and thedissemination of new ideas on the adaptation of multi-media systems to specific needs.

Collaboration with the Education and Culture DG hasstarted in the MOISE (Modello Organizzativo Internazi-onale di Servizi per bisogni Educativi speciali – Inter-national Organization Model for persons with SpecialEducational needs) project on education and disability.ISIS is helping a group of schools in the use of new in-formation technology tools. After validation by the finalusers, examples of best practices and design recom-mendations are selected and loaded onto the Internetin a format that can be accessed by all, irrespective ofthe required human-computer interface.

The team participates in an Inter-Service Group on Dis-ability where it helps identify user needs and translatesthem into recommendations and technical specifica-tions.

4.2.6 Multimedia Simulation System for the Disabled

P.C. CacciabueA scientific collaboration between ISIS and Istituto diRicovero e Cura a Carattere Scientifico (IRCCS) “EugenioMedea” was set up in order to promote a research pro-gramme aiming at a better integration of mentally dis-abled people in SMEs. The difficulty to recognise andreact to our own errors is a common experience. In par-ticular, this holds true for mentally disabled subjectsbecause of their low projective capabilities, their atten-tion deficit and the rigidity with which they apply reso-lutive schemes. Usually, during training they are taughthow to operate machines by simple automatic sequences.In order to devise a new training methodology and aninnovative way to deal with errors a quite innovativeapproach was taken.

A two-year project (1999-2001), financed by the ItalianMinistry of Health, was set up focusing on the develop-ment of a multimedia simulation system for improvingerror diagnostic skills in mentally disabled people. Thestarting point of the research was the developmentfrom scratch of a new classification for the analysis ofhuman-machine interaction hazards in the context ofmental disabilities [4.4]. The following step was thedevelopment of a simulation device that reproducesthe normal and abnormal (i.e., failures) functioning ofreal craft machines. The simulator will be embeddedinto a Computer Based Tool (CBT) and will be used fortraining purposes (Figure 4.13). In the second year ofthe project, likely scenarios to be played by the simu-

36

4.11 The project's co-ordinator is presenting the Voice

subtitling prototype to the European television

broadcasters at the Conference on Handicap and

Disability organised by RAI in Bologna. His speech is

on-line translated into subtitles and projected on the

wall screen.

4.12 A schoolteacher is using an automatic graphic system

based on Voice to help pupils with disabilities. As such

they may access games and tests in the same way as

their schoolmates.

lation system and guidelines for the design of the userinterface of the CBT were developed [4.5]. In particular,the latter subject was tackled in the context of a gradu-ation thesis carried out in collaboration with the Uni-versity of Turin - Department of Psychology.

4.2.7 Improvement of Error Diagnostic Skills

in the Disabled

P.C. CacciabueA graduation thesis [4.6] has been developed with theaim of studying and improving error diagnostic-skillsin young, mild mentally disabled people, in order toenhance their interaction with craft machines. Thisresearch activity is part of the collaboration with Istitu-to di Ricovero e Cura a Carattere Scientifico (IRCCS)“Eugenio Medea”.

A Computer Based Tool (CBT) has been developed andwill be dedicated to the professional preparation ofthese people; the CBT being part of a new trainingcourse. The analysis of users’ needs (teachers and pu-pils) allowed the definition of the software characteris-tics of the CBT. According to these needs, the CBT userinterface was designed as a series of “logical environ-ments”, that are mutually interconnected, each onefulfilling specific users’ goals. Five milieus have beenconceived within which the user interaction takesplace (Figure 4.14). Such a tool could be used within atraining course for a preparation to be held in a didac-tical laboratory, before the direct experience on the realmachines – the idea includes multiple appeals. Princi-pally, it provides a solution for safety-related problemsand, as a consequence, it allows pupils to experimentwith unusual situations, by stimulating at the sametime the development of their problem solving skillsand producing beneficial effects on their motivationand self-esteem. In addition, the use of software in alaboratory allows the collection of information to in-vestigate error modalities, through the possibility ofsystematic observations in a “controlled” environment.

It is important to point out that the main objective ofthe simulator training is not to transfer a specific andprecise competence, but rather to stimulate the develop-ment of disabled people’s reasoning skills. These canbe applied across different situations within the machineinteraction such as in other areas in everyday life. Hence,the most important objective in the development of thissoftware is not to create an exhaustive code capable ofaccurately reproducing reality, but to design a func-tional tool for teaching a behavioural style.


References

4.4 Sirtori, L., Pedrali, M., Pozzoli, U., Reni, G., Tacca, B.

– Towards the Implementation of a Multimedia Simu-lation System for Improving Error Diagnostic Skills inMentally Disabled People – presented at the Interna-tional Conference on Computer Helping People withSpecial Needs - ICCHP 2000 - July 17-21, KarlsruheUniversity, Germany, 2000.

4.5 Pedrali, M. Rossano, R. – A Computer Based Tool forEnhancing Human Machine Interaction of Mild Men-tally Disabled People – Proceedings of 4th HumanCentred Technology Postgraduate Workshop,Brighton, United Kingdom, 3-4 October 2000.

References

4.6 Rossano S. – Un contributo all’analisi dell’errore nelcontesto della disabilità psichica studio dell’inter-faccia utente di un software didattico – Master DegreeThesis in Scienza della Comunicazione, Università diTorino, Italy, to be presented 2001.

4.13 Screen shots of the demonstrator of the CBT (Control

Panel and Simulation environment).

4.14 General characteristics of the Computer Based Tool (CBT)

for training purposes of disabled people.

Start

Archive

End

Control panel

Theory Simulationenvironment Help

OVERVIEWJ.-C. Grossetie, H. Nguyen, A. Patak, J. C. Francescattihttp://3dips.jrc.it/

Activities within advanced integrated optical systemsconcentrate on the technologies needed for the appli-cations using 3D real time colour holographic synthe-sis. Diffractive optics (holograms, holographic opticalelements, computer generated holograms, diffractivegratings) differ from classical optics in the way that coh-erent light waves are simulated and used. Instead ofusing incoherent light technology that handles largecomplex images (e.g., in computer visualisation) newactive diffractive systems work with thousands of realor simulated coherent light waves; each system carry-ing much more limited information, but in the end, therecombination of the original real or simulated opticalwave gives a full 3D information. The diffraction phe-nomenon has become the primary tool for designingthe new passive or active optical systems, which canbe controlled almost in real time. This active real timecontrol is a fundamental advantage compared with theconventional optics and instruments that depend onentirely passive means.

The research concentrates on developing improved andup-to-date mathematical packages to simulate thesynthesis of true real colour holograms for realistic 3Dactive optical components and 3D reconstruction ofscenes, and for studying natural vision. Software hasbeen designed to carry out multi-stage progressiveand sequential 3D-hologram reconstruction whilstmaintaining very good optical quality. A large numberof applications (displays, storage, image synthesisand processing, etc.) can now be implemented usingthis advanced software. Since thin and volume holo-grams are mainly used for 2D and 3D static and dynamicrecordings, the first application and the most useful isthe 3D real time display, but other applications like inter-connects, memory data storage or optical correlatorsare equally important. Potential uses cover new activeoptical sensors; security and authentication products;real-time holography and active holographic opticalelements; advanced diffractive grating DOE; real timeholographic interferometry; active and passive holo-graphic optics; active and passive integrated optics;optical interconnects; opto-electronic devices used inadvanced informatics, passive optical elements, basicelements used in hard disk for data storage, RAM,DRAM memories, flash memories, new CD-ROM andDVD technologies.

5.1 MOVING FROM 2D DATA TO REAL 3DMEDICAL IMAGING

ISIS provides scientific support in validating and test-ing medical and health telematic systems and applica-tions, in particular concerning 2D and 3D imaging. Inthis context, common reference image sets and derivedimage libraries have been acquired and are availableto test and validate medical image algorithms and soft-ware applications. Starting from original and segmentedimages of the Visible Human Dataset, a reference libraryof 3D models of human anatomical structures and organshas been generated using the Visualization Toolkit. Togive a real 3D perception of the created models an in-novative auto-stereoscopic prototype screen is used.User-friendly generation of solid 3D anatomical modelsis addressed as well.

38

Medical Imaging – Data Systems5

5.1 Fourier and Fresnel Holograms Synthesis

Real-time Holographic Visualisation

Holographic Optical Element (HOE) Synthesis

Diffractive Optical Element (DOE) Synthesis

Innovative Holographic Flat Liquid Crystal Display

LCD Chip Design

Real-time Spatial Light Modulator (SLM)

Holographic Security & Authentication

5.2 Artefact example: some pixels belonging to the abdominal

aorta have been considered as belonging to the bone

tissues.


5.2 SCIENTIFIC AND TECHNICAL SUPPORT FOR MEDICAL IMAGES AND SYSTEMS

ISIS provides a coherent scientific and technical sup-port to explore the conception and implementation ofEuropean initiatives in the field of medical imaging andmedia interchange communication under the EuropeanHealth Policy. The objectives of the project are to study,develop and promote a European scientific and techni-cal framework for testing specific medical imaging andmedia communication systems in consultation withrelevant parties from the public and private sectorsat the European and national levels. In this context, afeasibility study of the research project called ImageGuided Orthopedic Surgery (IGOS) has been done.IGOS is a specific application of IOS (Image GuidedSurgery) in field orthopedic surgery. This field of sur-gery deals mainly with bones, spine and maxillofacialsurgery.

Main functionalities

• Make a 3D cephalometry

• Compute maxilla displacement

• Integrate mandible displacement

• Visualisation of maxilla and mandible displacement

• Export both displacements

Accidental trauma, dysmorphosys from birth maxilla or

mandible deformities.

5.3 Simultaneous rendering of bones, articulations and cartilaginous tissues of the Visible Human Male.

5.4 Image Guided Orthopaedic Surgery application.

OVERVIEWA. SaltelliThe focus at ISIS is in the domain of technology transferfor official statistics in support to EUROSTAT. Anothermajor commitment is devoted to user-driven methodo-logical support for economic and financial analysis, andthe competence being offered is in the domain of appliedstatistics and econometrics. This work supports theEuropean Investment Bank (EIB), the European CentralBank (ECB) and the Economic and Financial Affairs DG.

6.1 NETWORK BUILDING IN COLLABORATIONWITH EUROSTATA. Saltelli, R. Girardihttp://europa.eu.int/comm/eurostat/research/

http://www.jrc.cec.eu.int/uasa/prj-amrads.asp/

ISIS has an excellent collaboration with EUROSTAT.The main joint effort is in network building and tech-nology transfer. The technologies of interest are allthose potentially relevant for official statistics. ISISstimulates the creation of thematic networks and con-sortia on themes such as time series, confidentiality,metadata and others, and collaborates with EUROSTATin the organisation of international technology confer-ences. In this context ISIS also participates in theAMRADS project, an accompanying measure to buildnetworks in official statistics. Further more specificsupport is offered to EUROSTAT in the domain of timeseries analysis (section 6.4).

6.2 CONVERGENCE OF EMU ECONOMIESC. Planas, G. Fiorentini, G. Teyssierehttp://www.jrc.cec.eu.int/uasa/prj-busy.asp/

One of the most important achievements of the EU hasbeen the adoption of the Euro. Besides insuring stabilityof exchange rates within the Monetary Union, the mon-etary policy conducted by the European Central Bank(ECB) for the 11 States of the Monetary Union is a strongsupport for the co-ordination of the macroeconomicpolicy. It is thus an important matter to measure theconvergence of the Monetary Union economies (Figure

6.1). ISIS contributes to this issue with its ongoingwork for the standardisation of business cycle analysistools in Europe (BUSY project). BUSY is a 3-year collab-orative project (2000-2002) of the Fifth FrameworkProgramme. It is supervised by EUROSTAT and aims atmaking an organised set of tools for the statisticalanalysis of business cycles available to European eco-nomic and financial institutions. For this project, ISIScoordinates a consortium including the Spanish, theFrench and the Italian statistical offices. An interestgroup has been set up for this activity that includes theEconomic and Financial Affairs DG, the ECB, EUROSTAT,national statistical institutes and central banks. A majorconference on the convergence of the EU economies(Good Statistics for Good EU Governance) is foreseenfor the autumn of 2001, which will be jointly organisedby the Economic and Financial Affairs DG, EUROSTATand ISIS.

6.3 SHORT-TERM ECONOMIC INDICATORSA. Rossihttp://www.jrc.cec.eu.int/uasa/prj-flash.asp/

There is a crucial need for rapid estimates of the mainQuarterly National Account aggregates of the EuropeanEconomic and Monetary Union and of the EuropeanUnion. The target delay is 40-45 days after the end ofthe reference period. What central banks and the ECBneed is a coherent system that is able to help them toconduct short-term economic analyses and to takemonetary policy decisions. This would avoid the short-comings associated with the delays in availability ofthe official quarterly figures. ISIS is active in this areaand is collaborating with the German Institute of Eco-nomic Research, the Italian national statistical institute,the British NIESR and the National Bank of Belgium.

Applied Statistics6

Temporal Evolution of Economic Activities

Short-Term Indicators

Time series input data:Gross Domestic Product,

Consumer Price Index,Industrial Production Indices

Total Unemployment

Example of policy choiceto raise/reduce

interest rate

6.1 Recent advances in information technology now make it possible to access in real time a large number of economic time

series. This allows the derivation of short-term leading indicators that are crucial for public and private economic decision

making.

40


6.4 ECONOMIC TIME SERIES RELEVANT TOEMU AND EU AREASC. Planas, F. Archontakishttp://time-series.jrc.cec.eu.int/

Economic time series present themselves as time dep-endent jigsaws. These signals embody irregularities(e.g., an outlier due to strike) and yearly movements(e.g., Christmas spending frenzy, summer or Easterholidays), which hide the underlying long-term path(e.g., the trend of the economy). For monetary policymaking, it is important for Central Banks to have a clearpicture of the dynamic behaviour of these movements.In particular, extracting the long-term trend in con-sumer prices is crucial for the monitoring of inflation.ISIS contributes to this activity with its expertise ontime series analysis in collaboration with EUROSTAT.A methodological study was performed in 2000 for theECB on the seasonalilty in the European Industrial Prod-uction Indexes aggregated at the European MonetaryUnion level. An ongoing study for the Economic andFinancial Affairs DG is in the modelling of GDP and in-flation through estimation of Potential Output models.A comparison of EUROSTAT and ECB policy recommen-dations for seasonal adjustment has been produced atthe request of EUROSTAT as well as a proposal for aquality report on seasonal adjustment. On the sametopic a handbook jointly written by EUROSTAT and ISISis in preparation. ISIS also operates a web-based HelpDesk for statisticians involved in time series analysisand held a course on the subject on the premises ofthe ECB in Frankfurt in November 2000. According toEUROSTAT-A (Director P. Nanopoulos), from 2001 mostof EUROSTAT’s methodological activities of applicationof time series analysis to official statistics will be dele-gated to ISIS.

6.5 ANALYSIS OF FINANCIAL RISKF. Campolongo, A. Rossihttp://sensitivity-analysis.jrc.cec.eu.int/

Managing risk is a major concern for financial institu-tions worldwide, as shown by the very existence of theBasle Committee. As a major financing arm of the EU,the European Investment Bank (EIB) must excel in itsfinancial risk evaluation procedures in line with bestmarket standards. Mathematical models are devel-oped to evaluate the financial transactions that the EIBperforms daily and to construct a portfolio minimisingthe risk (be it market or credit risk). ISIS has discussedpotential areas of co-operation with the EIB both onmodel building and model assessment via uncertaintyand sensitivity analysis. Uncertainty and sensitivityanalysis are valuable tools in the practice of financialrisk assessment as they quantify the maximum riskincurred by a bank and identify what factors mostlycontribute to such a risk. In the perspective of Europeanintegration, balanced development, economic and social

cohesion of the Member States, and in the view of theenlargement of the EU, managing sophisticated financialrisk is a priority for the European financing institution.Two major results have been achieved in 2000:• The identification of concrete settings where uncer-

tainty and sensitivity analysis can be successfullyapplied to manage financial risk.

• The assistance provided to EIB for model building,e.g. the Hull and White one factor interest rate modelhas been studied and an implementation has beenprovided to the EIB, in addition to the interface withthe EIB’s operating environment.

6.6 ENVIRONMENTAL MODELS AND INDICATORSS. Tarantola, M. Ratto, N. Giglioli, J. Jesinghaushttp://esl.jrc.it/envind/dashbrds.htm/

http://www.jrc.cec.eu.int/uasa/prj-artemis.asp/

The sector also operates a number of activities relevantto sustainable growth and environmental policy, in thecontext of a growing support of ISIS to the EuropeanEnvironment Agency (EEA). These studies are under-pinned by ISIS’s combined experience in uncertaintyand sensitivity analysis, and modelling and indicatorsbuilding. The “Dashboard of Sustainability”, developedin co-operation with the International Institute for Sus-tainable Development (IISD) and the EEA, aims at com-municating complex sets of indicators in a format thatis accessible to decision-makers, stake-holders andthe general public. ISIS focuses on the theory of aggre-gating economic, environmental and social indicatorsto a Policy Performance Index (PPI), and of analysingthe linkages between these indicators. The main ob-jective of the “Dashboard” is to establish the “missinglink” between detailed indicator sets and non-expertusers without losing the information on the underlyingindicators, like for example, for the approximate 60 SDindicators promoted by the United Nations Commissionfor Sustainable Development and used by EUROSTAT.This project is also an important input for the Environ-ment DG in their preparation of Rio+10.

ARTEMIS is an international project financed by theEnergy and Transport DG where the experience fromdifferent emission calculation models is combined tobuild a harmonised methodology for transport emis-sion estimates, which serves to support EU air qualitypolicies. ISIS is one of the 34 partner institutions in theproject and its task is to make quantitative statementsabout uncertainties in emission estimates, which resultfrom complex error chains, and to make them availableto policy-makers, by considering the impact of differentscenarios, framing assumptions, and the effects of dif-ferent transport options. A first application was ad-dressed to the Frejus Tunnel area. COPERT-III (COmputerProgramme to calculate Emission from Road Transport)was extended to take into account uncertainties that

derive from meteorological data, fleet compositionand other model parameters. NOx, CO2, VOC and PM10pollutant emissions have been estimated for 2000 and2010.

IMPACT is a project financed by the Information SocietyDG. Natural fluctuations in the state of the environmentcan often overshadow the human impact. This makestools aimed at extracting anthropogenic signals frommore or less noisy data highly relevant for environ-mental assessment and policy making. The main objec-tive of IMPACT is to introduce new analytical tools (e.g.,global sensitivity analysis) that can facilitate assessmentof human impacts on the environment in the presenceof natural fluctuations in meteorological, hydrologicaland other conditions. Test cases worked so far includeair pollution and eutrophication problems at a regional/European scale.

The demand of industry in terms of complex problemsolving has resulted in ISIS engaging in applied statis-tical work that targets specific industrial needs. A col-laboration has been established between AGIP and ISISaimed at applying ISIS's expertise in the conceptualand computational environment used by engineers forthe analysis of the oil-potential of sedimentary basins.The objective of the work is to improve the success rateof drilling by combining petroleum system modelling(AGIP) with global uncertainty and sensitivity analysis(ISIS).

6.7 DISSEMINATIONA. Saltelli, F. Clement http://time-series.jrc.cec.eu.int/

http://sensitivity-analysis.jrc.cec.eu.int/

http://europa.eu.int/comm/eurostat/research/

2000 was an intense year of dissemination for the sec-tor. The Help Desk forum TSAOS for time series analy-sis was launched as well as a forum for sensitivityanalysis. The publishers John Wiley & Sons publishedISIS’s book on sensitivity analysis (Figure 6.2). A sem-inar was held in Brussels in June to communicate to theEIB, the ECB, the Economic and Financial Affairs DG andothers the most recent activity of the group, especiallyin short term analysis (BUSY, FLASH) and financialmodelling. Altogether the sector maintains approxi-mately 100 pages of web material related to its work andis moving into the field of XML technology.

42

6.2 “Too often modellers do not realise that sensitivity

analysis is an essential part of the model building

process. This volume has a didactical value showing how

sensitivity analysis is often useful – and sometimes

essential – to complete the model building process and to

interpret results properly. It guides the reader through an

array of different approaches, illustrating in a generally

clear fashion the specificity of the different techniques to

different problem-settings” (Amazon.com review).

43


Pastorelli R., Tarantola S., Beghi M.G., Bottani C.E.,

Saltelli A. – Design of surface Brillouin scatteringexperiments by sensitivity analysis – Surface Science.468, 37- 50, 2000.

Saltelli A., Tarantola S., Campolongo F. – Sensitivityanalysis as an ingredient of modelling – StatisticalScience, 15(4), 377-395, 2000.

Saltelli A. – Un portale al servizio degli statistici per iltrattamento delle serie storiche – Giornale delSISTAN, 15, 2000.

Crosetto M., Tarantola S., Saltelli A. – Sensitivity andUncertainty Analysis in Spatial Modelling Based onGIS – in Agriculture, Ecosystems and Environment,71-79, 2000.

Tarantola S., Giglioli N., Jesinghaus J., Saltelli A. – GlobalSensitivity Analysis to assess the quality of aGIS-based model – Proceedings of ACCURACY 2000,Amsterdam, July, 2000.

Sobol I.M., Tarantola S. – Global sensitivity analysis forComplex Simulation Models – Proceedings of MCS2000, International Conference on Monte CarloSimulation, Monte Carlo, June 18-21, 2000.

Tarantola S. – Quantifying Uncertainty Importance whenInputs are Correlated – In MP Cottam, RP Pape, DWHarvey and J. Tait Eds., Proceedings of ESREL 2000,Edinburgh Scotland UK, May 15-17, 2000 1115-1120.

Saltelli A., Chan K., Scott M., (Eds.) – Sensitivity Analysis –John Wiley & Sons publishers, Probability andStatistics series. The following Chapters are (co-)authored by ISIS: §1, Saltelli A., What is SensitivityAnalysis?, 3-13. § 2, Campolongo F., Saltelli A.,Sørensen T., Tarantola S, Hitch hiker’s guide tosensitivity analysis, 15-47. § 3, Campolongo F. andSaltelli A, Design of experiment, 51-63. § 4,Campolongo F., J. Kleijnen, Andres T., Screeningmethods, 65-80. § 8, Chan K., Tarantola S., Saltelli A.,Sobol’ I. M. Variance based methods, 167-197. § 12,Scott M., Saltelli A., Sørensen Tine, Practicalexperience in applying uncertainty and sensitivityanalysis, 267-274. §13, Draper D., Saltelli A.,Tarantola S., Prado P., Scenario and parametricsensitivity and uncertainty analyses in nuclear waste

disposal risk assessment: the case of GESAMAC,275-292. § 14 Planas C., Depoutot R., SensitivityAnalysis for Signal Extraction in Economic TimeSeries, 293-307. § 15, Tarantola S., Pastorelli R.,Beghi M.G., Bottani C. E., A dataless pre-calibrationanalysis in solid state physics, 311-327. §18,Campolongo F., Saltelli A., Comparing Different SAMethods on a Chemical Reaction Model [KIM],355-365. § 19, Zaldivar J., Campolongo F., Anapplication of sensitivity analysis to fish populationdynamics, 367-383. § 20, Tarantola S., Jesinghaus J.,Puolamaa M., Global sensitivity analysis: a qualityassurance tool in environmental policy modelling,385-397. § 22, Saltelli A., Fortune and future ofSensitivity Analysis, 421-426. Appendix, Chan K.,Scott M., Andres T., Software for Sensitivity Analysis:A Brief Review, 451-464 (2000).

Girardi R. – Introduction to sensitivity analysis ofconditional forecasting, a variance based applicationto econometrics – EEA 2000-6, Istituto di EconomiaPolitica, Studi e Quaderni, Serie di Econometriaapplicata, Universitá Bocconi, Milano, Italy.

Saltelli A., Tarantola S. – Sensitivity Analysis: aprerequisite in model building? – In MP Cottam,RP Pape, DW Harvey and J. Tait Eds., Proceedings ofESREL 2000, Edinburgh Scotland UK, pp. 1121-1126.15-17 May 2000.

Fischer B., Planas C. – Large scale fitting of regressionmodels with ARIMA errors – Journal of OfficialStatistics, v.16, n.2, pp 173-184, 2000.

Chan K., Saltelli A., Tarantola S. – Winding stairs - Asampling tool to compute sensitivity indices –Statistics and Computing, 10, 187-196, 2000.

Campolongo F. Rossi A., Saltelli A. – The role of sensitivityanalysis in financial modelling – Proceeding of the 5thInternational Conference on Probabilistic Safety As-sessment and Management (PSAM5). November2000, Osaka, Japan, 2000.

Depoutot, R., Planas, C. – Finite approximation to linearfilters and the monitoring of revisions – ComputationalStatistics, 15(1), 25-30, 2000.

Cyber-security, Knowledge Access & Assessment, the Fight against Fraud

S. Funtowicz, A.Pereira, S. Corral

7.1 SCIENCE AND GOVERNANCEhttp://alba.jrc.it/science-governance.html

The many recent debates on science-related issues (forinstance, technological and environmental risks) showthat the reliability of policy-relevant scientific informa-tion can no longer be assumed. Unlike most other proc-esses and products, scientific information does not havean established quality assurance programme, beyondthe informal practice of scientific peer review and refe-reeing of basic research and project proposals. Increas-ing attention is being paid to scientific aspects of policyor administrative decisions taken at European levels.Certain features of the process, including the poten-tially irreversible consequences of decisions, the un-certainties involved and the pressure on scientists toproduce a particular result, indicate the need for a newrelationship between science and governance. Thisshould incorporate greater openness and more partici-pation in the process from stakeholders.

7.2 QUALITY ASSURANCE OF POLICY-RELATEDRESEARCHhttp://alba.jrc.it/pedigree.html

In the appraisal of strategic policy issues, scientific in-formation is but one of the inputs. It is supplementedand interpreted by knowledge of local contexts, as filter-ed through political processes. The analytical workneeds to proceed in recognition of social controversy,uncertainties, and plurality of decision criteria. The ap-propriate methods include multi-criteria aids, institu-tional analysis and deliberative techniques. Informationtechnology plays an essential role in the design anddevelopment of appropriate tools that capture thecomplexity of policy and decision exercises. Their powerenables them to avoid an exclusive dependence onquantification: they permit the construction of “spacesof negotiation” where all the stakeholders can see theirconcerns expressed and where potential compromisescan be explored.

Moreover, a programme of knowledge assessment isessential when information that is used as an input topolicy making is complex – deriving from severalsources or evaluated by several criteria – uncertainand even disputed. There is a need for an agreed andcredible set of procedures or guidelines by which thereliability of information can be assessed, and alsoassent established among all stakeholders in thisprocess. In particular NUSAP and Pedigree schemescan provide elements on quality, legitimacy and relia-bility of several levels of information processing forpolicy making. Thus the Pedigree scheme has beenconceived as an exploring tool through which threemain elements of decision-making processes – the

information, the decision tool and the role of the analyst– can be assessed. The community involved in thedecision-making process such as experts, politiciansand citizens defines criteria and attributes implementedduring this exploration. The Pedigree Scheme has beenrecently implemented in a case study of environmentalplanning dealing with atmospheric pollution issues.

7.3 INFORMATION TECHNOLOGIES ANDKNOWLEDGE ASSESSMENT

Information technologies have a great and growingpower to provide information, quickly, cheaply, in largequantities to many users and to integrate differenttypes and sources of knowledge that characterise manycomplex problems. Nevertheless, the technologiesthemselves do not guarantee the quality of that infor-mation. The development of a sustainable informationsociety requires public trust in the technology and theinformation it provides. Many issues must be resolvedby the providers and the users and regulators of thetechnologies. These include the reliability of the infor-mation, the protection of the privacy, security and civilliberties of its users, and its accessibility by all sectorsof society, as well as the openness, transparency andlegitimacy of the technology and its governing institu-tions. Then the information technologies will realisetheir promise as the means for providing a forum forbroad and effective citizens’ participation, enabling thegrowth of a sustainable information society.

44

DecisionProblem

Uncertainty IgnoranceLack of

Information

High StakesUrgent

DecisionsDifferent

Perceptions

DecisionProcedure

PublicPolicy

Knowledge Assessment Methodologies7

The ArtLab is a multi-media laboratory founded by the

Knowledge Assessment Methodologies sector to deploy

web-based multi-media products for the information socie-

ty. The ArtLab activity aims at responding to current issues

on science and governance and knowledge assessment, in

particular on the wider access and participation of the civil

society in policy and decision-making. This is done by ex-

ploring the opportunities of new information communica-

tion technologies, but by putting the emphasis on the place

of individuals in cyber-contexts.

The application of a knowledge assessment approachenables the systematic analysis, appraisal and assess-ment of the use and reliability of appropriate knowledgefor policy processes. It is essential to explore ways tomake that information available in a transparent, openand reliable way; this has to be done with the audiencesfor whom the information is intended. Of particular rel-evance here is the development of Internet-based in-teractive multi-media information technology and thecreation of tuning contexts by means of deliberativeprocedures and other knowledge assessment mecha-nisms, so that the final design of these systems andpitfalls of usage are fully understood by all concerned.

7.4 VISIONS - INTEGRATED VISIONSFOR A SUSTAINABLE EUROPEhttp://alba.jrc.it/visions.html

The VISIONS project has been implemented in 3 differ-ent regions in Europe; the JRC/ISIS being responsiblefor the Venice region. The set of four images of the futureof Venice in 2025 was prepared by means of multimediamaterials, providing insights into systemic interactionsamongst different dimensions of sustainability withconnections to European scale scenarios being devel-oped by the co-ordinator of the project, thus showingin an interactive way, the unfolding of the imaginedvisions of the future. These scenarios have been pro-posed through a social process to stakeholders andcitizens in Venice. Early versions of the scenarios weretuned for their contents – put into context – and visual-isation by taking into account the opinions of the par-ticipants in the social process. Later, the scenarios wereproposed to the participants to explore them as toolsto engage the society in a debate about policy options,empowerment, responsibility and governance issuesto attain desired futures.

7.5 GAS - GREENHOUSE GASES METER,SCENARIO EXPLORER AND DAY PLANNERhttp://alba.jrc.it/gas.html

GAS is much more than a simple software implementa-tion project of a tool that relates lifestyles with green-house gas emissions and possibly with climate change:this tool embeds new concepts of the communicationof science to non-scientific audiences and therefore itreflects concerns on a number of quality issues: (1) in-formation: reliability of sources, data uncertainty andignorance; (2) calculation core: uncertainty of model

ling, transparency, assumptions; and (3) user interfaceto scientific issues: intuitive interface, “vocabulary”,tangible, transparency.

GAS is an example of the sort of new tool that is beingdeployed by an information society that tries to engageall citizens in the problems that affect us all. Its aim isthe empowerment of all citizens about a global issueand their assuming responsibility for it in order to en-hance the processes of governance. The validation ofthe software was financially supported by the Enter-prise DG and consisted not only of the scientific basisof the calculation core, but also of the appropriatenessof the interface to be used by non-specialised users.The final version of GAS is available for 5 differentcountries in Europe: France, Italy, Portugal, Spain andthe UK including data for each of those countries. GAS isdistributed as a stand-alone tool in the form of a CD-ROM.

7.6 GOUVERNehttp://alba.jrc.it/gouverne.html

The project GOUVERNe (Guidelines for the Organisation,Use and Validation of Information Systems for Evaluat-ing aquifer Resources and Needs) aims at the develop-ment and pilot implementation of a user-based andscientifically validated Decision Support System (DSS)for the improved management of underground waterresources at the catchment and sub-catchment levels.The role of the envisaged DSS is not to generate a “best’or ‘optimal” management choice in a technocraticfashion, but rather to furnish a validated scientific sup-port for debate and deliberation by decision-makers,stakeholders and citizens allowing for compromises,generation of novel options and, to the greatest extentpossible, co-operative conflict resolution. The emphasisin GOUVERNe is on a process of stakeholder-based app-raisal and management means that decision supportis understood not only as the acquisition, scientificvalidation and organisation of the information, but alsoas procedures for effective exploitation of this informa-tion by users. This is another case of creating qualityassured interfaces between science and society. ISISis involved not only in the creation of the actual multi-media interface between the modelling deployed bythe DSS but also in the knowledge assessment process,i.e., the scientific quality assurance and the creation ofa tuning context where these tools are evaluated fortheir fitness by end-users.


7.1 Image composed from drawings taken from Moebius’s

Venice Celeste.

But, Marco Polo... the cities you aretelling me about seem to be the samecity... Maybe your stories are differentways of looking at it. Perhaps a visionof Europe.

46

Fight against Fraud88.1 ANTI-FRAUD INFORMATION MANAGEMENT

OverviewT. BarbasWork in information technologies during 2000 focusedprimarily on improving our early warning capacity andon developing new open source intelligence capabilities,including new developments in the area of informationextraction from texts. This is expected to imply advancesin both strategic and operational intelligence capacitiesfor our customers and partners. At the same time, oureffort to raise awareness about fraud control techniquesamong participants in anti-fraud networks was signifi-cantly intensified. As in previous years, mixed funding,internal and external, was used to finance the activities.While the primary beneficiary of our work was the Euro-pean Anti-Fraud Office (OLAF), other beneficiaries in-cluded the Agriculture DG and Member State specialistagencies. Finally, a strategic collaboration with theRegional Policy DG was launched in the area of informa-tion technologies applied to financial management forthe Structural Funds.

8.1.1 Early Warning Analysis

T. Barbas, C. Moussas, A. Burger, A. VarfisIn early warning, analysis of relevant data sources con-tinued from previous years. This included a study tointegrate IRENE (reported frauds and irregularities data-base of OLAF) fraud case data into a generalized archi-tecture for sectorial data marts of particular interest toOLAF’s intelligence and operational strategy depart-ment [8.1]. In December 2000, we concluded work in datamodelling, transformation and harmonisation workaimed at the development of a data warehouse for ex-port refunds. The beneficiary of this work was Clear-ance of Accounts of the Agriculture DG. The tools usedfor data navigation, visualization and reports were Busi-nessObjects and WebIntelligence (of BusinessObjects,France). We also invested some effort in migratingexisting consolidated import trade data for textiles(the result of previous years’ work) to a web-compatibleenvironment using SAS/IntrNet software (of SAS Insti-tute, USA). Finally, a commitment was made with theExternal Relations DG’s financial management depart-ment for a strategic study of relevant information tech-nologies to be applied to the management of paymentdata, both at the aggregate and the detailed account-ing levels.

Again in the area of early warning, the software appli-cation EWS-GIS (Early Warning System for Transit) wasfinally handed over to the customer (OLAF) [8.2]. Resultsachieved in 2000 included usability and functionalityimprovements, as well as a prototype for web-basedaccess of maps visualized by the system. The systemwas actually field-tested during the Joint Customs Sur-

veillance Operation WAMBA in Alicante (1-16 June 2000)and was also used for training participants in newanalysis techniques in a special preparatory meetingfor the same operation in Toledo.

The Neural Networks and Machine Learning (NNML)Laboratory of ISIS supported this line of activitiesthrough the continued development of a software proto-type to detect outliers in large databases. As a result ofthe latest enhancements, the software is now also ofinterest to domain or problem experts with no particularprogramming or computer skills.

8.1.2 Name Data Cleaning and Information Portals

T. Barbas, A. GargTwo important projects were carried over from 1999.Both were financed by OLAF, at least in part. The first isDOCTUM-V2, a powerful, completely rewritten versionof an earlier prototype for cleaning structured data in-side files or databases – like names and addresses ofcompanies, or locations [8.3-8.4]. DOCTUM-V2, like itspredecessor, uses a number of approximate matchingtechniques. With respect to the prototype, the softwarecan now export directly clean data into third party soft-ware for visualisation or (network) analysis. In addition,the operations necessary to group similar informationitems together became more powerful and the user inter-face was significantly improved. It should be added,however, that DOCTUM-V2’s matching power, as definedby the set of methods and algorithms it contains, rem-ained unchanged with respect to the earlier prototype.

The second important project carried over from 1999 isIIIMS-V2. [8.3]. More than an information portal in anti-fraud, IIIMS-V2 is meant to be a kind of knowledgebase, a virtual forum where investigators and analystscan share, assess and exchange their knowledge aboutgood sources of information. The system’s true powerwill only be revealed if the user population is extendedto include national partners over an anti-fraud network(linked via an Extranet or similar mechanism). With res-pect to its predecessor (IIIMS), IIIMS-V2 runs as a webserver and also includes facilities for personalised in-formation capture and filtering from the web. At thetime of writing of this report, development of IIIMS-V2is 80% complete.

References

8.1 Moussas C., Barbas T. – Integrating IRENE Data intoOLAF’s Data Warehouses – Technical NoteNo. I.00.54, April 2000.

8.2 Burger A. – EWS-GIS: a GIS for the Transit of Sensi-tive Goods Registered in EWS – Project Report for theperiod 1999-2000, July 2000.

8.1.3 Open Source Intelligence

J. Perdigao, A. Garg, T. BarbasIn the area of open source intelligence capabilities, aftera first-level requirements’ analysis and technology review[8.5], our work focused on the development of a soft-ware prototype intended to improve current monitor-ing capacity for the movements of containers, based oninformation available over the Internet. One contextwhere this is useful is when customs investigators arelooking for evidence involving past activity of specificcontainers (itineraries, loading or unloading events etc.).Another context is pro-active risk analysis to better tar-get physical inspections. Technology-wise, this dev-elopment combines web engineering for gatheringand filtering information, with data warehousing fortransformations and consolidation of data and, finally,data analysis and visualisation. The validity and rele-vance of the approach was confirmed by many membersof European maritime customs intelligence forumsduring a number of presentations by JRC-ISIS staff. Adecision will be taken in early 2001 as to whether toproceed with a pilot project.

8.1.4 Language Engineering

R. Steinberger, S. Scheer, J. HagmanWork in Language Engineering continued from previousyears along the lines established for the present Frame-work Programme. The high-level long-term goals arefighting the information overflow and overcoming thelanguage barrier on behalf of our partners, with a focusto giving cross-language access to information hiddenin large amounts of multilingual text. During 2000,advances were made in the multilingual retrieval of pot-entially relevant documents, in the extraction of differ-ent information aspects from documents (keywordsand named entities such as geographical references)and the visualisation of the contents of documents(profiles for individual documents and document mapsfor collections) – Figure 8.1. Efforts to put together amultilingual tool set were continued with componentseither developed in-house or first purchased and thenintegrated. The Modus Operandi project, which had thegoal of analysing and comparing fraud-related docu-ments in multilingual collections, was successfullycompleted [8.6 - 8.11]. An evaluation of commerciallinguistic software was carried out following a requestof OLAF [8.12]. Finally, an important result this yearwas the capability to gather information on Internetcrime automatically. This was a deliverable of projectOSILIA (Open Source Intelligence Library on InternetAbuse). The solution involved the use of web crawleragents that search on-line newspapers for potentiallyrelevant articles [8.13].


References

8.3 Checchi E., Valerio G., Garg A., Ferigato C. – Intools-2:Functional Specifications – Technical NoteNo. I.00.134, November 2000.

8.4 Mastrangelo G., Garg A. – DOCTUM Version 2.0, UserManual – Special Publication No. I.00.119, November2000.

References

8.5 Garg A. – Automatic Intelligence gathering from theweb: Pre-Analysis, (interim project report) – May2000, Technical Note No. I-00.137.

8.1 Document profile and document map

(map is produced using ThemeScape by Cartia Inc.).

8.1.5 Dissemination

T. Barbas, J. Perdigao, R. Steinberger,A. Garg, S. Scheer, J. Hagman

There were many occasions this year where project res-ults were disseminated to anti-fraud partners, goingbeyond the usual presentations of results to custom-ers. For example, in addition to the Joint Customs Sur-veillance Operation already mentioned above, ISISpresented work in automatic intelligence gathering totechnical meetings of Marinfo-North and Marinfo-South(maritime customs intelligence forums), as well as theOpen Source workshop of EUFIS (the European Forumfor Intelligence Specialists). As for work in LanguageTechnology, ISIS presented its findings also in aca-demic conferences like [8.14-8.15].

8.2 DATA ANALYSIS AND RISK ANALYSISFOR ANTI-FRAUDS. Arsenis, F. De Amicis, V. Alexandrou

In 2000, ISIS continued work on a number of data analy-sis and risk analysis issues pertaining to the exploitationof data on irregularities and frauds committed againstthe budget of the EU and reported by Member Statesto the European Commission. This work was conductedon a recent complete download of the IRENE databasein which the cases are stored. IRENE is a database ofOLAF, which is the primary customer of the project.

In 1999, an examination of cases reported by MemberStates against the European Agricultural Guidance andGuarantee Fund (EAGGF) had been done at the requestof OLAF, on the basis of subject matter knowledge todifferentiate between frauds and irregularities. In 2000,this work was extended to cases reported against thetraditional own resources and the Structural funds[8.16]. Descriptive statistics on the number of casesand amounts established were calculated and tabulat-ed by Member States and robust statistical tests wereapplied to bring out Member States reporting signifi-cantly large or small cases. Zero amounts establishedappear in the data surprisingly frequently: 23% of thecases against the EAGGF and 20% of the cases againsttraditional own resources. The sector developed andapplied a procedure of data-mining two-way contin-gency tables and characterised such cases by MemberState, and status variables for the case. The groupingof all types of irregularity into two categories (fraudsand other irregularities) is meaningful when cases arecharacterised appropriately. To take into account possi-ble heterogeneities in reporting practices, [8.10] alsoderives and tabulates signals either in the frequencyof reporting cases or in the amounts established forcombinations of Member States and types of irregularity.

In order to carry out the work reported in [8.16], exten-sive referencing had to be done to data stored in fieldsnot previously retrieved from the cases downloaded.This included a study of the structure of the templates inwhich cases are reported. The programming work to re-trieve data from the templates is documented in [8.17].

The numerous tables of signals on two-way categoricaltables presented in [8.16] were produced in two stages,first programming in SAS and then processing SASoutputs in VBA (see Figure 8.2). This is one of a numberof similar SAS-VBA batches for presentation of knowl-edge extracted from databases by using statisticalmethods. Such batches are expected to simplify verylaborious, cumbersome and repetitive tasks and makepresentation of results aesthetically appealing.

48

References

8.14 Steinberger R., Hagman J., Scheer S. – Using Thesaurifor Information Extraction and for the Visualization ofMultilingual Document Collections – in Proceedingsof the Workshop on Ontologies and Lexical KnowledgeBases, Sozopol, Bulgaria, September 2000.

8.15 Hagman J., Perrotta D., Steinberger R., Varfis A.

– Document Classification and Visualization to Supportthe Investigation of Suspected Fraud – Working Notesof the Workshop on Machine Learning and TextualInformation Access (MLTIA) at the 4th EuropeanConference on Principles and Practice of KnowledgeDiscovery in Databases, Lyon, September 2000.

References

8.6 Steinberger R., Hagman J., Barbas T. – ModusOperandi Final Project Report – Summary andConclusions. Technical Note No. I00.88.

8.7 Steinberger R. – Software Solutions to Overcome theLanguage Barrier – Technical Note No. I.00.91.

8.8 Steinberger R., Hagman J. – Commercial KeywordIdentification and Clustering Software – TechnicalNote No. I.00.90.

8.9 Hagman J. – An Implemented Cluster Analyser forDocuments and Their Indexing Terms. – TechnicalNote No. I.00.106.

8.10 Hagman J. – Some Ways of Visualising Results ofCluster Analysis – Technical Note No. I.00.107.

8.11 Steinberger R. – The Free Text Field of the IRENEDatabase – Technical Note No. I.00.89.

8.12 Steinberger R. – Evaluation of Document ManagementPartners’ Linguistic Software – Internal reportrequested by OLAF, December 2000.

8.13 Scheer S., Steinberger R., Valerio G. – A Methodologyto Retrieve, to Manage, to Classify and to QueryOpen Source Information – Results of the OSILIAProject (Open Source Intelligence Library for Internet

Abuse). – Technical Note No. I.01.016.

Report [8.16] below contains partially confidential dataand will not be disseminated freely.

8.3 ADVANCED STATISTICS FOR THE CLEARANCE OF ACCOUNTSS. Arsenis, A. Karvounaraki, T. Vogiatzi

2000 was the second year of execution of the AdvancedStatistics for the Clearance of Accounts project. Theproject aims to provide a valid scientific basis for thechoice of appropriate sampling plans and estimationmethods to be adopted and practised in conductingaudits for payments made to beneficiaries of theCommon Agricultural Policy (CAP) in order to estimateoverpayment in payment populations. The primarycustomer for the project is the Agriculture DG and thesecondary customer is the European Anti-Fraud Office(OLAF).

After having examined payment and audit data fora compliance audit conducted by the Clearance ofAccounts unit and having reported on such data andproposed a new estimator for overpayment [8.18], the

project team focused on the current state of practice ofcontrols conducted by the certifying bodies in theMember States. The reports of two certifying bodieson two paying agencies were studied with an interestin the extent to which previous auditing experiencecan be of use in current audits. As a result of the study,the sector concluded that particular emphasis has tobe placed on how to standardise the control work andthe presentation of results deriving from it, the choiceof control parameters, i.e., audit risk, materiality andexpected error to define sample size, the estimationand testing procedures used, and the definition oferrors. Another issue of importance is the extensionof methods and techniques applied meaningfully byone certifying body and in one budgetary line to otherpaying agencies. Finally, we quantified the extent towhich beneficiaries confirmed by the certifying bodiesto have been involved in irregularities or frauds wereindependently known to have been associated withsimilar practices in the past. Results of this work werepresented in a seminar to customer staff. The final reporton this work [8.19] incorporates input and commentsfrom the participants in the seminar.

In order to formulate the present state of knowledgeon overpayment errors, the project team scanned allfinancial clearance decisions and compliance decision200/216/EC under the reformed Clearance of Accountsprocedure and compiled a four-way table of correc-tions done per Member State, sector, budgetary posi-tion and year [8.20]. In the future, amounts correctedwill also presented as fractions of advances paid. Thiswork will cover the five remaining compliance deci-sions to maintain updated summaries of the presentstate of corrections.


References

8.16 Arsenis S., Alexandrou V. – Characterization of IRENECases with Particular Reference to Irregularity Typeand Established Amount – ISIS/RMDS/268v4, mainreport 95 pp., appendices 428 pp. (2000).

8.17 De Amicis F. – Data Retrieval from the Free Text ofIRENE3 Download – ISIS/RMDS/ 362v2, main report11 pp., appendices 48pp. (2000).

8.2 Data mining cross-classification of cases reported

against the EAGGF by Member State (masked) and cases

being fraud attempts (“TF”) or “other”.

(c) Signal extraction statics for THRESH=1, alpha=0.05

Count_0 FS Ncompo Nsignals Ncatx0 Ncatxs Ncaty0 Ncatys

8495 0.80 28 12 14 12 2 2

(b) Signals extracted

PS OTHER TF CX_S CX_0XX1 36 36 60XX4 1548 1548 1779XX5 537 537 767XX6 752 752 855XX7 11 11 14XX8 395 395 412XX9 94 94 160XX10 1542 1542 1558XX11 182 182 708XX12 489 489 505XX13 160 160 172XX14 1058 1058 1104C_YS 5784 1020 6804C_Y0 6964 1531 8495

(a) Raw cross-classification

OTHER TF TotalXX1 24 36 60XX2 127 21 148XX3 192 61 253XX4 1548 231 1779XX5 230 537 767XX6 752 103 855XX7 3 11 14XX8 395 17 412XX9 66 94 160XX10 1542 16 1558XX11 526 182 708XX12 489 16 505XX13 12 160 172XX14 1058 46 1104Total 6964 1531 8495

50

The highest quality audit data on payments made tobeneficiaries of the CAP are those obtained by theCourt of Auditors DAS (Déclaration d’assurance) annualexercises. The project team has looked into the char-acteristics of the tainting populations and expects toobtain access to the complete samples.

Reports [8.18-8.19] contain partially confidential dataand will not be disseminated freely.

8.4 IDEA (IDENTIFICATION ELECTRONIQUEDES ANIMAUX)C. Korn, O. Ribó

The objective of the project IDEA (Identification Elec-tronique des Animaux) [8.21-8.23] is to assess the feas-ibility of electronic identification for farm animals andto validate identification systems in real field conditions(Figure 8.3). Approximately one million animals fromthree species (440,000 cattle, 490,000 sheep, 30,000goats) in six EU Member States (France, Germany, Italy,The Netherlands, Portugal and Spain) are to be identi-fied by applying three different types of electronictags: injectable transponder, ruminal bolus and eartag– Figure 8.4. The performance of electronic identificationdevices and the necessary organisational structure ina real EU livestock scenario are to be evaluated. Theanalysis will determine whether electronic identifica-tion is a feasible system to trace animals individuallyfrom birth to the slaughterhouse. It is important toidentify the appropriate technologies for electronictags and reader types for full-scale implementation inEU livestock.

The IDEA project (1998-2000) was launched by the Agri-culture DG and has been recently extended until theend of 2001 in accordance with the current Regulation1760/2000 on the identification and registration of bov-ine animals, which states in article 4 that not later than31 December 2001, the Council shall decide on

the possibility of introducing the electronic identifica-tion system in the EU livestock. ISIS is responsible forthe technical and scientific support, the testing andcertification of electronic identification devices, quali-ty control of equipment, definition and establishmentof the central database, data transmission and record-ing during the project, and global evaluation of resultsobtained.

The design and development of the IDEA Central Data-base was finished in 1999. The JRC IDEA database wasimplemented and tested and has been operationalsince the beginning of 2000, when the data stream fromthe IDEA participants to ISIS started. During 2000, allparticipants finished the test phase of the data trans-mission and some IDEA participants began to send realdata, which was analysed at a syntactic and semanticlevel, validated and entered into the IDEA database. Adocument, which defines the statistical analysis to beperformed with the IDEA data, was prepared. The soft-ware programming of the statistical tools to analysethe IDEA data commenced at the end of 2000. During2000, six new electronic identification devices (elec-tronic tags, and portable and static readers) were cer-tified in the TEMPEST (Thermal, Electromagnetic andPhysical Equipment Stress Testing) laboratory. Thus,since 1996, 82 pieces of equipment have been certifiedand 343 certificate updates have been produced.Complementary research was conducted to study theinfluence of various parameters such as animal speed,electromagnetic interferences, etc. on the reading effi-ciency of the tags. Quality control of the tags used bythe participants also started in 2000 and 8 samplingswere conducted with 8 different participants in order:• to check that all tags used by the contractors have

the Laboratory Acceptance Certificate and correspondexactly with the reference tags described in the cer-tificate.

References

8.18 Arsenis, S., Alexandrou, V., De Amicis F. – On PaymentData for Export Refunds and Results of an Auditconducted after Risk Analysis Considerations –ISIS/RMDS/278 (2000).

8.19 Arsenis, S., Karvounaraki A. – On Reports of Certify-ing Bodies with Particular Reference on Harmonisa-tion in Reporting – ISIS/RMDS/402 draft (2000).

8.20 Vogiatzi T. – Corrections imposed by Financial andCompliance Decisions under the New Clearance ofAccounts System – ISIS/RMDS/403 draft (2000).

8.3 Reading station for electronic animal identification.

• to check the performances of the tags used by thecontractors (in real conditions) for the IDEA project.A comparison between normal production units versuslaboratory-tested samples for certification was alsocarried out.

The electronic identification of animals within theIDEA project began in September 1998. During 2000,IDEA participants supplied data from 275,000 cattle,400,000 sheep and 25,000 goats electronically identi-fied applying ruminal bolus (n=525,000), electroniceartags (n=150,000) and injectable (n= 25,000), whichrepresent 75% of these animals in the IDEA project. Itis expected that in 2001 this will reach a total of 1 millionanimals to be identified within the IDEA project.

A preliminary analysis of the results indicates thatelectronic eartags and ruminal bolus presented thelowest mean percentages of losses compared to inject-able transponders. The retaining rates of ruminal bolusand electronic eartags are higher than commonly obs-erved retaining rates in plastic eartags. These prelimi-nary results appear to indicate that electronic taggingusing ruminal boluses and electronic eartags seems tobe an efficient and reliable method for livestock identi-fication. However, this conclusion is drawn only from afraction of the data to be collected. Detailed data onloss, breakage, electronic failures by animal type, tag-ging and reading data, equipment, people, etc. shallbe taken into consideration when further analyses willbe performed with more data.


References

8.21 Korn Chr. – List of Certificates of LaboratoryAcceptance for the IDEA Project – Version 2.0 of15/03/00. Sa-Ve-Tech Unit, ISIS Institute, JRC Ispra (I),Technical Note no. I.00.45, March 2000.

8.22 Ribó O, Cropper M., Korn C., Poucet A., Meloni U.,

Cuypers M, De Winne P. – Preliminary results on elec-tronic identification in sheep and goat in the IDEAproject – Book of abstracts of 51st Annual Meeting ofthe European Association for Animal Production(EAAP). The Hague, The Netherlands, 21-24 August2000. Book of abstracts No. 6 (2000), p. 318.

8.23 Ribó O, Cropper M., Korn C., Poucet A., Meloni U.,

Cuypers M, De Winne P. – Animal Tagging andTraceability activities in the framework of the fightagainst fraud – Proceedings of the Seminar “LatestDevelopments in Livestock Identification andTraceability” (CD-Rom, MS Office Format, 6 pp.).Meat and Livestock Commission (MLC), Milton Keynes(UK), 16 February, 2000.

8.4 (a) Injectable transponder; (b) ruminal bolus;

and (c) electronic eartag.

8.5 NEW TECHNOLOGIES FORMONITORING FISHING VESSELSI. Shepherd, N. Kourtihttp://intelligence.jrc.cec.eu.int/fish/index.htm

The European Union’s Common Fisheries Policy man-ages fisheries for the long-term benefit of fishing com-munities and consumers. The objective is to regulatethe amount of fish taken from the sea so that youngfish can reproduce. Management measures include:• total allowable catches (TACs) which limit the total

amount of a species that can be caught in a given areain a given time;

• quotas which determine how much of a given TAC isallocated to each country;

• technical measures which might restrict the type offishing gear to be used or the season in which fish-ing is permitted in order to protect young or non-targeted fish.

Regulations to monitor and check these measures aredecided at an EU level and implemented by the MemberStates. These regulations include the obligation for allvessels more than 24 metres in length – over 4,000vessels in the EU – to install satellite-based navigationand communication equipment. The position of thevessel is then transmitted on an hourly or two-hourlybasis to a Fisheries Monitoring Centre in its flag state.The flag state then sends it on to an equivalent centrein the coastal state. The system is at present still beingimplemented – the transmission from vessel to flagstate works well in a number of Member States, butthere is some work to do for the re-transmission to thecoastal state. Clearly vessels that do not carry this on-board equipment or whose on-board equipment is notworking cannot be monitored in this way.ISIS has been helping the Fisheries DG assess the pos-sibilities of using space-borne Synthetic Aperture Radar(SAR) to detect and possibly identify these vessels.The study has covered three test areas:• The Flemish Cap – in international waters in the North

West Atlantic (summer 1999).• The North Sea (spring 2000).• The Azores (spring 2000).

In general, the challenge is to detect targets against anocean background. The noise from the background islowest, and therefore the detection easiest, if the SARbeam is polarised horizontally and if it has a large inci-dence angle, i.e., the satellite is low on the horizon.There is a trade-off between fine resolution and largearea coverage. Of the commercial satellites currently inoperation, the Canadian RADARSAT satellite offers themost appropriate options. In the studies it was foundthat its SCANSAR mode, although relatively coarse inresolution, was best for vessel detection because itcovered a large sea area – 300 x 300 km. Most opera-

tional vessel detection systems rely on a human operatorexamining an image, but ISIS felt that an operationalsystem would need a more automatic system and dev-eloped a PC-based software including land-maskingand fast geo-referencing.

The Flemish Cap studies showed that nearly all thevessels could be detected on the images – only on veryrare occasions were the sea conditions such that thebackground noise was too high. The detected positionswere then compared against VMS reports obtainedfrom the Icelandic, Norwegian, Portuguese and Spanishfisheries inspectorates. It was found that 72 % of thevessels whose position was given in the VMS reportscould be identified – the reason for non-identificationusually being a too infrequent reporting interval forVMS and the resulting uncertainty in the vessel’s posi-tion at the acquisition time. The average discrepancybetween positions determined from the image and posi-tions determined from VMS was 0.3 nautical miles. Thiswas considered an outstanding result considering allthe potential inaccuracies – georeferencing, VMS un-certainties and doppler shifts due to vessel velocity. Areport was delivered to the Fisheries DG and the resultssummarised at meetings of EU Fisheries inspectors.

The studies in the North Sea and Azores were in collab-oration with the UK’s Defence Research EvaluationAgency (DERA) who run a satellite ground station atWest Freugh in Scotland. VMS positions were suppliedby the Dutch, Danish (for the North Sea) and Portuguese(for the Azores) authorities. The aims were to examinethe potential of SAR imagery in areas of heavy maritimetraffic (North Sea) and in remote areas where control isdifficult and to determine whether the results could begiven to inspectors in real time. In a number of trialruns, times of 2.5 hours were obtained from image acq-uisition to completion of the analysis. The limitingfactors were the processing at the ground station andthe transmission of the 140Mb processed image fromthe ground station to the analysis centre. Options forimproving the speeds were examined and costed. Anumber of new difficulties was highlighted – the real-time images are not so well geo-referenced as thoseprocessed some days later because of inherent orbituncertainties and small vessels requiring lower thresh-old parameters, which may give some false alarms.Based on these results an outline design for an opera-tional system was developed. The vessel identificationrate in the North Sea was found to be about 90%, whilein the Azores it was less than 10% – the reason beingthat the Azores fishing vessels are made of fibreglasswith different backscattering properties to those ofsteel.

52


Kourti N., Shepherd I., Pavlakis P., Hoffmann V.

– SAR-imagery for fishing vessel detection –Final report October 2000, Technical NoteNo. I.00.120. 468, 37- 50, 2000.

Kourti N., Shepherd I., Brock D., Moesel S., Griffith P.,

Martin D. – Ship Detection for Fisheries Monitoring –Backscatter, Fall 2000.

8.5 The detected vessels are in red and the detected oil rigs

are in blue. The green points show VMS position reports.

Some effort was started to see whether the new high-resolution satellite, IKONOS, could provide useful in-formation for vessel identification. Both panchromaticand multi-spectral images of an area in the Adriatic Seawere analysed. Although effectiveness may be limitedby the small area covered and the disturbance of clouds,the ability of this satellite to provide evidence for fish-ing activity should not be underestimated. Furtherstudies are required to find a methodology to coupleIKONOS imagery with other monitoring and controltools to fully exploit its possibilities. IKONOS imageryas well as other optical imagery can be used in areassuch as the Mediterranean where the majority of thevessels are below 24 m and mainly made of fibreglassor wood.

No one method or technology can provide all the infor-mation that authorities need under all circumstances.It has become clear that what is needed is an integrationof information of tools and technologies to provideknowledge. Work is continuing on behalf of the FisheriesDG, and in close collaboration with authorities in theMember States and their suppliers of advanced moni-toring and communications technology, first to furtherdevelop and assess these tools and technology andsecondly to help with their integration into completepackages.


54

9. Safety and Emergency Management of Man-made and Natural Hazards 56

9.1 Major Accident Hazards Bureau (MAHB) 56

9.2 Natural and Environmental Disaster Information Exchange System (NEDIES) 58

9.3 Management of Natural and Technological Hazards in Pre-accession Countries 60

9.4 Competitive Activities and the Major Accident Hazards Bureau 60

9.4.1 Assessment of the Uncertainties in Risk Analysis of Chemical Establishments 609.4.2 Safety Management Systems for Major Accident Hazards in Small and Medium Enterprises 609.4.3 Network on Hazard Assessment of Highly Reactive Systems 61

9.5 Risk Monitoring and Emergency Management 61

9.6 Software Package for Civil Protection Emergency Planning and Management 62

9.7 Safety of Industrial Processes and Human Factors 62

9.8 Thematic Network on Safety Assessment in Waterborne Transport 63

9.9 Automotive Collision Avoidance System 63

9.10 Monitoring of Landslides 64

9.11 Detection of Oilspills 65

10. Decision Support for Risk and Emergency Planning 66

10.1 Enhancing Scientific Underpinning to Agri-environment Policies 66

10.2 MATADOR 68

10.3 UTOPIA 68

10.4 The CommonGIS Project 69

10.5 The MURST project 70

11. Aviation Safety 72

11.1 European Co-ordination Centre for Aviation Incident Reporting Systems (ECCAIRS) 72

11.2 Human Factors in Accident Databases 73

11.3 Competitive Activities on Human Factors in Aviation Safety 74

11.3.1 Aircraft Dispatch and Maintenance Safety (ADAMS) 7411.3.2 Development of a Method for Air Transport Safety Improvement through

Quantitative Risk Evaluation (DESIRE) 7511.3.3 Advanced Cross-Communication Environment providing Support Services to Dispersed

Maintenance and Technical Support Engineers (ACCESS-maints) 7511.3.4 Advanced Integrated Training in Aeronautics Maintenance (AITRAM) 7611.3.5 Aviation Safety Targets for Effective Regulation (ASTER) 7611.3.6 Collaborations 77

12. Seismic Protection for Civil and Cultural Heritage 78

12.1 Seismic Protection for Civil and Cultural Heritage 78

12.2 Activities concerning Fibre Reinforced Composites (FRC) 79

12.2.1 Upgrading and Repair of Reinforced Concrete Elements 7912.2.2 Application of Composites to Seismic Resistance of Reinforced Concrete Buildings 80

12.3 Seismic Assessment of Masonry Structures 80

12.4 Further Developments in Dynamic Control of Earthquake Engineering Facilities 82

12.5 Active and Semi-Active Control in Civil Engineering (ACE and CASCO) 82

12.6 Vulnerability Assessment of Bridges (VAB) 83

12.7 Anchorages in Concrete under Dynamic Loading 84

12.8 The ICONS and SAFERR Research Networks 84

12.9 Project LADIB (Laboratorio Dinamico Industriale della Basilicata) 86

12.10 Tunnelling Project 86

12.11 3D Reconstruction 87

NATURAL AND TECHNOLOGICAL RISKSAND EMERGENCIES

13. Safety in Transient Structural Dynamics 88

13.1 Numerical Modelling - EUROPLEXUS 88

13.1.1 Industrialisation of FEM Advanced Software 8813.1.2 Report on Phase 1 of the EUROPLEXUS Agreement with CEA 8813.1.3 Benchmark Tests 9013.1.4 Documentation 9013.1.5 Data Structure Modernisation 90

13.2 Structural Aspects in Transport Safety 90

13.2.1 Structural Crash Safety Enhancement by Precision Impact Tests 9013.2.2 Road Barrier Deformation Tests 9213.2.3 Pedestrian Safety 93

14. Nuclear Safety 94

14.1 Severe Accidents 95

14.1.1 Melt Coolant Interactions 9514.1.2 Structural Integrity of Reactor Vessel in Severe Accidents 9614.1.3 Hydrogen Risk 9714.1.4 Source Term 97

14.1.4.1 Phébus Experimental Programme 9714.1.4.2 Phébus Network 9914.1.4.3 PHEBEN2 10014.1.4.4 EVITA 10014.1.4.5 ICHEMM 100

14.2 Thermal Hydraulics 101

14.2.1 ASTAR 10214.3 Safety of New Designs 103

14.4 Harmonisation 103

14.4.1 Probabilistic Safety Assessment 10414.5 Dissemination and Collaboration 105

14.5.1 CERTA 10514.5.2 ASTERISM2 10614.5.3 Enlargement 106

14.6 Spin-off Activities 106

14.6.1 Project Epsilon 10614.6.2 Sub-Aero 10714.6.3 Lung 10714.6.4 Predicting Internet Traffic 107

55Natural and Technological Risks and Emergencies

56

OverviewA.C. LuciaIn this cluster of projects (a cluster that also includesthe projects presented in Chapters 10 and 11) the JRC isacting as a centre dedicated to the development andvalidation of scientific advice at the EU level and to theestablishment of a scientific and technical reference inthe development of a system approach to the manage-ment of natural and technological hazards. The projectssupport the implementation of policies aimed at theprotection of individuals, infrastructures and societyand are in accordance with the guidelines expressedby the European Research Area initiative.

The new attention paid by citizens to decisions thatmay have consequences on their health, safety, qualityof life or the state of the environment, the ever increas-ing complexity of technological systems and extensionof highly industrialized areas and the still unsatisfyingdegree of preparedness to face disasters and mitigateconsequences are the main driving forces of theseactivities.

ISIS efforts are focused on:• Accident reporting and analysis, which is done at the

European scale with the necessary independencefrom national and private interests and which assuresthat experience is shared among all Member Statesand data kept confidential when required;

• Support to the conception, preparation and imple-mentation of European directives;

• Use of up-to-date information and communicationtechnologies offering reliable and powerful meansfor collection, assessment, dissemination and correctand timely exploitation of data, information and know-ledge.

• Helping candidate countries align more quickly withthe European Union’s “acquis communautaire” infields reflecting pre-accession process priorityneeds.

• Methodological developments, validation and ap-plications, mainly performed in the framework oflarge international projects (typically shared-costactions) and third party works for national or regionalauthorities (such as the SIMAGE project for the ItalianMinistry for the Environment).

9.1 MAJOR ACCIDENT HAZARDS BUREAU (MAHB)J.S. Duffield, A.C. Lucia, N. Mitchison,C. Kirchsteiger, M. Christouhttp://mahbsrv.jrc.it

The process industry covering the petrochemical,chemical, agrochemical and pharmaceutical industriesis a major economic force, which employs millions ofpeople and generates billions of Euros in shareholdervalue and tax revenues. It is more than twice the size of

the global market for telecommunications equipmentand services and accounts for 7 % of global incomeand 9 % of global trade. The safety and public accept-ance of this industry is therefore of paramount impor-tance to our society.

The Major Accident Hazards Bureau (MAHB) gives in-dependent scientific and technical support to the Euro-pean Commission in the field of EU policy on the controlof major hazards involving dangerous substances andthe prevention and mitigation of major accidents. Thismost notably concerns the successful implementationand monitoring of the “Seveso II” Directive 96/82/ECfor the control of major hazards in fixed installations.Furthermore, in order to fulfil its information exchangeobligations towards the Member States, the Commis-sion established the Major Accident Reporting System(MARS) and the Community Documentation Centre onIndustrial Risks (CDCIR), which are managed and main-tained by MAHB.

The principal customers of MAHB are the EuropeanCommission, particularly the Environment DG, and allplayers concerned with process plant safety includingthe legislative and regulatory process – these customersinclude national and local authorities, industry, researchorganisations, safety consultants and trade unions.

Although the transposition of the “Seveso II” Directiveinto Member States’ national law should have takenplace by February 1999, many Member States encoun-tered difficulties in achieving this goal, due in no smallpart to the fact that this was a completely new direc-tive. The directive introduced significant changes topolicy related to the control of major hazards, such assafety management systems, formalised risk assess-ment, land-use planning etc., and it is these changesand their consequences which were the focus of muchof MAHB’s work in 2000.

One of the major activities of MAHB results from thefact that Member States are legally obliged to notifythe occurrence of major accidents to the EuropeanCommission, and significant effort is devoted to theanalysis of these accidents and to the extraction anddistribution of lessons learnt. To this end, MAHB oper-ates and maintains an accident reporting system calledMARS. Using this system, Member State authoritieshave a local data logging system, which they use toprepare accident reports. These reports are then sub-mitted in electronic form to the central MARS systemrun by MAHB where advanced analysis tools havebeen developed to facilitate in-depth analysis of theaccidents notified. To enhance transparency and tofulfil our obligation regarding information to the publicmuch of this data is made available through the MAHB

Safety and Emergency Managementof Man-made and Natural Hazards9

57

website. Furthermore, in recognition of the value andusefulness of our accident reporting system MAHB hasbeen approached by the OECD and the UN/ECE organi-sations with a request to design and manage similaraccident reporting systems for their purposes. Collab-oration agreements have been drawn up between ISISand the two international organisations, and during2000 pilot versions of the OECD and the UN/ECE MARSversions were completed. In this way MAHB has becomeprobably the world centre for major accident reporting,and furthermore by having a uniform data structure forthe information reported it will greatly facilitate theanalysis of the root causes of accidents. Closely coupledto the accident reporting system and forming an inte-gral part of the software is a GIS-based application inwhich the geographical component of risk associatedwith industrial plant is developed. This windows-basedapplication, the “Seveso Plant Information RetrievalSystem” (SPIRS) has the goal of containing all relevantdata on industrial plant throughout Europe which hasto comply with the directive. It will enable the Commis-sion to perform hazard potential and risk categorisa-tion assessments on a uniform basis. Complementingthe MARS/SPIRS software is a dangerous substancesdata management system (DMDMS) which holds specificinformation, such as physical properties and qualify-ing quantities on all the named substances and genericcategories in the directive.

The Community Documentation Centre on IndustrialRisk (CDCIR) now holds over 3,700 documents and is avery important source of knowledge on process safetyand industrial risk. This library holds, in addition todocuments conventionally published, a wide range of“grey literature”, such as company reports, codes ofpractice, accident reports, safety studies, etc. A newweb-based facility has been extended in the year 2000to holding indexing information, abstracts, authors,keywords etc., for all these documents thereby ena-bling on-line access to the CDCIR through the MAHBwebsite.

MAHB, in close collaboration with the EnvironmentDG, runs a number of Technical Working Groups(TWGs). In most cases the primary objective of thesegroups is to develop guidance to ensure a coherentand correct implementation of the Seveso II Directive.It was a major achievement that, as a compendium tothe directive and prior to the transposition of the direc-tive into Member States’ national law, a full set of“guidance” documents was completed, published andwidely distributed by MAHB. All these documents canbe viewed and downloaded from the MAHB website.During 2000, TWG7 “Substances Dangerous to theEnvironment” and TWG8 “Carcinogenic Substances”have completed their work. These two working groups

were set up at the request of the European Parliamentand Council to investigate the appropriate thresholdlimits for substances classified as being dangerous tothe aquatic environment, and to identify substancesand the appropriate threshold limits for substancesthat can cause cancers from a single exposure, such asthat following a major accident. The recommendationsof the working groups have been published [9.1-9.2]and they will be used as the scientific basis for a futureamendment to the directive.

In an attempt to promote good practice and seek a uni-fied implementation of the directive, MAHB and theEnvironment DG have initiated a “Mutual Joint VisitsProgramme” (MJV). This activity aims at facilitating in-formation exchange among the Member State inspectorsby allowing a host country to present and discuss theirinspection systems with fellow inspectors from otherMember States. This has proven to be very successfuland to date visits have taken place in the Netherlands,Ireland, Germany, Finland and the UK.

In addition to its institutional activity, MAHB has beeninvolved in a number of competitive activities. During2000, these have included the ASSURANCE shared-costaction project that focuses on assessing the uncertain-ties associated with the most commonly used riskanalysis methodologies for assessing major industrialhazards and the SMMARTEN shared-cost action projectthat focuses on safety management systems for smalland medium sized enterprises and the HARSNET the-matic network that focuses on hazard assessment ofhighly reactive systems.


9.1 Consequences of the explosion at the fireworks factory in

Enschede.

MAHB has a long tradition in organising internationalseminars on themes related to industrial risk. In 2000,in view of the transposition of the Seveso II Directive,an important international conference was organisedin London entitled “The Implementation of Seveso II”,in which delegates from competent authorities, indus-try research organisations and academia were able topresent their views on the worth, practicalities andfunctioning of the directive. Following the fireworkaccident in Enschede, (Figure 9.1) in which 21 peoplewere killed and 944 were injured, a more restrictedseminar was organised in Marseille, linked to the com-mittee meeting of the competent authorities. The resultsfrom this meeting together with a follow-up meeting in2001 will form the basis for an additional amendmentto the directive.

Risk assessment plays a central role in the directiveand MAHB together with the nuclear safety unit of ISISorganised an international workshop on “Promotionof Technical Harmonisation on Risk-Based DecisionMaking” [9.3] where experts drawn from all types ofactivities and industries discussed the possible needand perspectives of developing an internationallyaccepted generic “standard” for risk-based decisionmaking.

The growing importance of the Internet has been rec-ognised and MAHB maintains an active and dedicatedwebsite (http://mahbsrv.jrc.it/) from which documen-tation, software tools, MAHB publications and relevantinformation can be readily downloaded by interestedparties. A selected list of MAHB publications for 2000is given in [9.4-9.12].

9.2 NATURAL AND ENVIRONMENTAL DISASTERINFORMATION EXCHANGE SYSTEM (NEDIES)A.G. Colombo, A.C. Lucia, C. Theofilihttp://nedies.jrc.it/

NEDIES (Natural and Environmental Disaster Informa-tion Exchange System) is a project concerned with theanalysis of natural and environmental disasters thathave occurred in the EU Member States. It has beenpromoted by the Environment DG and is under devel-opment at ISIS.

The main aim of the project is to produce “lessonslearnt reports” by type of disaster. Lessons learnt reportsenable the interested organisations in the MemberStates to gain from the analyses of the experiences inother countries. Moreover, it is expected that MemberStates can take into consideration this beneficial infor-mation for planning future activities so as to mitigatethe harmful consequences of disasters. The processinvolves the following steps:

58

References

9.1 Christou M. (Ed.) – Substances Dangerous for theEnvironment in the context of Council Directive96/82/EC (Seveso II) – Final Report of TWG 7, EUR19651 EN (2000).

9.2 Christou M. (Ed.) – Carcinogenic Substances in thecontext of Council Directive 96/82/EC (SevesoII) –Final Report of TWG 8, EUR 19650 EN (2000).

9.3 Kirchsteiger C., Cojazzi G. (Eds.) – Promotion ofTechnical Harmonization on Risk-Based Decision-Making – Proceedings of Workshop, Stresa, Italy,May, 2000 S.P.I.00.63

9.4 Duffield J.S., Lucia A.C., Mitchison N., Kasamas H.

– Land recovery and man-made risks. A perspectivefrom the EU accession countries – Journal of HazardousMaterials, Vol. 78, pp. 91-103, 2000.

9.5 Christou M., Mattarelli M. – Land-use planning in thevicinity of chemical sites: Risk-informed decisionmaking at a local community level – Journal ofHazardous Materials, Vol. 78, pp.191-222, 2000.

9.6 Kirchsteiger C. – Availability of Community LevelInformation on Industrial Risks in the EU –Transactions of the Institute of Chemical Engineers,Vol. 78B, pp. 81-90, March 2000.

9.7 Contini S., Belezza F., Christou M., Kirchsteiger C.

– The use of geographic information systems in majoraccident risk assessment and management – Journalof Hazardous Materials, Vol. 78, pp. 223-246, 2000.

9.8 Christou M., Lauridsen K., Amendola A., Markert F.,

Kozine I. – ASSURANCE: ASSessment of Uncertaintiesin Risk ANalysis of Chemical Establishments, inPSAM 5 – Probabilistic Safety Assessment andManagement, K. Furuta and S. Kondo (Eds.),[Proceedings of the International PSAM 5 Conference,Osaka, Japan, 2000], Universal Academy Press, 2000.

9.9 Kirchsteiger C., Cojazzi G. – Promotion of TechnicalHarmonization on Risk-Based Decision Making, inPSAM 5 – Probabilistic Safety Assessment andManagement, K. Furuta and S. Kondo (Eds.) idem.

9.10 Kirchsteiger C., Duffield J.S. – Recent Developmentsin European Commission Tools to Manage IndustrialRisks in Europe – in Proceedings of the InternationalConference on Process Industry Incidents, CCPS, Oct2000, Orlando, Florida, USA.

9.11 Mengolini A., Christou M., Contini S., Ludovisi G.

– Rischio di incidente rilevante nel trasporto multi-modale di sostanze pericolose, con riferimento aipunti di interfacciamento tra i modi, in Atti dellaConferenza Nazionale “Valutazione e Gestione delRischio” - VGR2k – Pisa, Italy, Oct. 2000.

9.12 Kirchsteiger C., Rushton A. – Human ErrorContributions to Major Industrial Accident Scenarios– in Proceedings of the 18th ESReDA Seminar on“Risk Management and Human Reliability in SocialContext”, Karlstad, Sweden, June 2000.

Natural and Technological Risks and Emergencies 59

• identification of the disasters to be analysed andcollection of the relevant documentation (usuallythrough an ad-hoc meeting);

• analysis of the information collected and productionof a lessons learnt report;

• dissemination of the documents produced.

The disasters to be considered in the analysis are select-ed according to their relevance, taking also into accountthe availability of the needed endorsed information.The analysis is structured according to the manage-ment phases of a disaster: prediction, prevention, pre-paredness and response. Particular consideration isgiven to the information supplied to the public before,during and after the event. The role of mitigation in thethree phases is investigated.

In 2000, five lessons learnt reports have been complet-ed, which deal with avalanche disasters, flood disasters,catastrophic earthquakes, train accidents and tunnelaccidents, respectively.• Avalanches: six events were considered, four of

which occurred in 1999: Gargellen (Austria), northernpart of the Swiss Alps (Switzerland), Chamonix(France) and Aosta Valley (Italy), and two occurredin 1995, in the west fjords of Iceland. The main lessonslearnt from those disasters are reported in [9.13].

• Floods: three flood disasters were selected for theanalysis. They are: the 1993 flood in Limburg (Bel-gium), the 1995 high water emergency in the Nether-lands and the 1997 summer flooding of the Oder inGermany. The lessons learnt from those floods arediscussed in [9.14].

• Earthquakes: the analysis concerned four events,which occurred in Greece: the earthquake in theAttica area (2000) that caused more than one hundredand forty fatalities, and the earthquakes in Kalamata(1986), Grevena - Kozani (1995) and Egio (1995). Therelated lessons learnt are reported in [9.15].

• Train accidents: six accidents were investigated.These include the ICE train disaster in Eschede (Ger-many), which occurred in 1998 and caused morethan one hundred fatalities, the 1999 accident in thePaddington (UK) railway station and the 1991 crashin the railway station at Melun (France). The ammoniaderailment in Kävlinge (Sweden) that resulted in theevacuation of 9,000 people for twenty-four hours,the 1995 train accident, which occurred in KlostertalMason Valley (Austria) and the 1993 train accidentin Naestved (Denmark) were also considered in theanalysis. The main lessons learnt from these accidentsare presented in [9.16].

• Tunnel accidents: nine accidents were examined.These consist of the 1999 fire in the Mont Blanc tunnelbetween France and Italy that caused 38 fatalities;two accidents that occurred in Austria of the same

year (Tauern tunnel and Pfänder tunnel), which alsocaused several fatalities; the fire in the Channel tunnelbetween UK and France, which occurred in 1996;and another five accidents that, fortunately, did notprovoke the loss of lives. Lessons learnt from thesenine accidents are reported in [9.17].

To collect and store the endorsed information on natu-ral and environmental disasters, and to disseminatethe information collected and produced, a website(http://nedies.jrc.it/) has been designed and imple-mented (homepage shown in Figure 9.2). The informa-tion contained in the website is arranged into twoparts: “Open Information” and “Password-ProtectedInformation”. Non-registered users are able to readthe “Open Information” that consists of a short des-cription of the background of the project and the maincontents and achievements of the two phases. Theyare also able to read and download the main reportsissued by the project team. Authorised users can putinto the database, protected information on disastersand read all the information stored. A password isrequired, however. Authorised experts are able to searchfor disasters, specific documents (i.e., documentsrelated to a specific disaster input in the system) andgeneric documents (i.e., documents containing genericinformation on natural disasters and their management),and report generic documents. Nominated representa-tives are able to report and search for disasters andboth types of documents (generic and specific).

9.2 The NEDIES homepage.

References

9.13 Colombo A.G. (Ed.) – NEDIES Project, Lessons Learntfrom Avalanche Disasters – Report EUR 19666 EN(2000).

9.14 Theophili C. (Ed.) – NEDIES Project, Lessons Learntfrom Flood Disasters – (2000).

9.15 Theophili C. (Ed.) – NEDIES Project, Lessons Learntfrom Earthquake Disasters in Greece – (in press).

9.16 Colombo A.G. (Ed.) – NEDIES Project, Lessons Learntfrom Recent Train Accidents – Report EUR 19667 EN(2000).

9.17 Colombo A.G. (Ed.) – NEDIES Project, Lessons Learntfrom Tunnel Accidents – (in press).

9.3 MANAGEMENT OF NATURAL ANDTECHNOLOGICAL HAZARDS INCANDIDATE COUNTRIESA.C. Lucia, J.S. Duffield, N. Mitchison

This activity started in the second half of 2000, in thecontext of the JRC Enlargement Programme, on thegrounds of the experience acquired and the toolsdeveloped in the institutional project described insections 9.1 and 9.2. Its main objective is the designand creation of compatible regional and national centralinformation systems for supporting authorities in themanagement of risk and emergency situations due tonatural and technological hazards and to the existenceof heavily polluted land sites. Candidate (Central andEastern Europe) countries have to face challenges andproblems of a large dimension and complexity arisingfrom a wide range of natural and technological hazards.Furthermore, quite a serious threat to the safety andhealth of citizen is constituted by the existence of landpolluted by industrial spoil, military activities and im-proper handling of hazardous wastes: actions for landrecovery or for reduction of risk have to be undertaken.

The envisaged steps of the activity are:

1. Creation of a collaboration structure in the countriesconcerned.

2. Collection of data and information on• sites that are potential sources of major accident

hazards;• sites that are potential sources of environmental

pollution;• sites where there is soil or groundwater pollution

(from current or past activities;• past accidents and events and related reports and

lessons learnt;and creation of the relevant geo-referenced data-bases and information systems.

3. Creation of risk maps.4. Development and application of methods for the

risk-based screening of sites and installations andfor the prioritisation of the interventions.

The year 2000 activity (started after the summer) hasbeen devoted to the identification of official key contactpersons in the 10 pre-accession countries, throughcontacts with the embassies and ministries, and to thecreation of a network of technically involved persons.A second part of the activity has been the preparation,organisation and planning of an international workshop“Management of Natural and Technological Hazards”to be held in Ispra on 14th-16th March 2001, with theparticipation of delegates from these countries.

9.4 COMPETITIVE ACTIVITIES AND THE MAJORACCIDENT HAZARDS BUREAUJ.S. Duffield, M. Christou, A. Mengolini, M. Fiori

9.4.1 ASSessment of the Uncertainties

in Risk Analysis of Chemical Establishments

(ASSURANCE)

The ASSURANCE project began in 1998 and will becompleted in the spring of 2001. It aims to improve theunderstanding of the uncertainties associated with therisk analysis of major industrial hazards by assessingthese uncertainties and the way that they can affectthe final outcome of risk studies. This is achieved byassessing the state-of-the-art (including approaches,techniques, models, and assumptions) of risk analysisin the European Union; analysing the sources of uncer-tainty present in risk assessment and their contributionto the uncertainty and variability of the overall results;and comparing the approaches used to deal with theseuncertainties. The method adopted was that a referenceinstallation was analysed by different teams independ-ently of each other and the results compared. Specialexercises/case-studies (with well-defined commonassumptions and based on the reference installation)were carried out and intermediate results from eachphase of the risk assessment procedure are beingcollected and compared to identify which phase – andwhich source – contributes most to the variation of theoverall results. The implications of these uncertaintiesin policy issues, such as land-use planning, are alsobeing examined. Seven teams from throughout Europehave performed these case studies and ISIS, as techni-cal co-ordinator, is evaluating these results.

9.4.2 Safety Management Systems for

Major Accident Hazards in Small and

Medium Enterprises (SMMARTEN)

The SMMARTEN project recognises the specific needsof small and medium sized enterprises (SMEs) work-ing with dangerous substances in their efforts to fulfiltheir legislative obligations. It is a project funded underthe CRAFT programme and was the first of its type forthe JRC. The main objective of the project was to bringtogether a number of companies with research organi-sations and to identify priority issues among the severalelements of a safety management system (SMS) whereguidance and support to SMEs is most needed. Recog-nising the limited resources that SMEs possess to for-malise the way they manage safety, a central featureof the project was to develop an easy-to-use method-ology to support the companies in the implementationof an effective safety management system (a majorrequirement of the Seveso II Directive). In addition, theproject consortium has tried to identify the functionalcharacteristics of a dedicated SME network, which willestablish contacts between those enterprises interested

60


in improving their approach to safety management sys-tems and interested service and knowledge providers.This network is expected to provide support on riskmanagement and control issues to a wider range ofSMEs operating in the European Union.

9.4.3 Network on Hazard Assessment of

Highly Reactive Systems (HARSNET)

HARSNET is a Brite-Euram thematic network on Hazardassessment of Highly Reactive Systems – it has over35 members that are drawn from the major industrial,academic and research organisations throughout Eur-ope. The main objective of the network is to transferthe safety knowledge, procedures and best availablepractices from large companies and research centresto small and medium sized chemical industries. It followsthe urgent need to introduce hazard assessment tech-niques for reactive chemicals into the normal workingprocedures of SMEs. Most of the expensive and special-ised experimental techniques commonly employed insuch studies are not usually easily available to suchconcerns, and with this in mind, new guidelines are beingproduced including a step-by-step assessment proce-dure based on a literature survey, simple predictivecalculations and laboratory based testing in readilyavailable equipment. Such procedures are being incor-porated into a software package (HarsMeth) which isdesigned to reveal the occasions (typically 5 %-10 % ofall reactions) where more sophisticated tests shouldbe performed by external laboratories or specialisedexperts.

Other activities of the network include the productionof HarsBook, which is based on the accumulated andshared experience of the partners and describes the bestmethods currently available for predicting, assessingand minimising exothermic reaction hazards. HarsBaseis also being developed, which is a collection of data-bases and links to chemical reaction safety related datathat can be easily be accessed by SMEs through theInternet. Finally, an important spin-off from the net-work is the formulation and promotion of new researchprojects to support the methodologies underpinningHARSNET. In this context, a shared-cost action project(AWARD) has recently been granted that concentrateson the early detection of runaway reactions using chaostheory and the safe disposal of reaction products. ISISis playing an active role.

9.5 RISK MONITORING ANDEMERGENCY MANAGEMENTJ.P. Nordvik, A.C. Lucia, D. Chavaudra

This was the first year of the 3-year framework agree-ment with the Italian Ministry for the Environment forthe establishment (at the national level) of an integrat-ed system for environmental monitoring, risk manage-ment, and environmental end technological emergencymanagement. The major objectives of the project,called SIMAGE (Sistema Integrato per il MonitoraggioAmbientale e la Gestione del rischio industriale e delleEmergenze), are:• the creation of harmonized air-quality networks for

the industrial areas of Brindisi and Taranto, includ-ing the integration and optimisation of existing air-quality networks, the installation of new monitoringstations and air pollutants instruments and the defi-nition of quality-control procedures and associatedcontrol laboratories;

• the design and establishment of a national co-ordination centre for environmental information ex-change concerning air, water and soil quality, net-worked with the major existing Italian risk areas,in particular, the areas of Brindisi, Taranto, PortoMarghera, Priolo-Augusta, Gela, Milazzo, Genova,Ravenna and Livorno-Piombino;

• the development of a pilot system for the monitor-ing and control of the transportation of dangeroussubstances via road, railway and sea, including abenchmark exercise of tracking and mobile commu-nication technologies, and the realization of localsystems in Brindisi, Taranto and Porto Marghera fortraffic control and emergency management;

9.3 One major objective of the Framework Agreement

called SIMAGE established between the JRC and the

Italian Ministry for the Environment is the develop-

ment of an integrated information exchange network

that covers the main existing Italian risk areas.

Ispra

Marghera

Ravenna

Genova

Livorno

Piombino

Roma

Brindisi

Taranto

Milazzo

Augusta/PrioloGela

Palermo

62

SIMAGE started in Autumn 1999 with ISIS being incharge of the design, realisation and effective estab-lishment of the integrated system, as well as of theoverall project co-ordination. Results in 2000 includedthe activities of:• identification and preparation of user requirements

at national and local levels;• high-level design and architecture definition of the

Brindisi and Taranto air-quality monitoring net-works, of the national overall environmental infor-mation exchange network, and of the pilot systemfor the monitoring and control of the transportationof dangerous substances;

• organisation and running of the major call for tenderrelated to air-quality monitoring networks and asso-ciated systems.

ISIS also continued to manage the monitoring systemsthat it installed during the previous years in the threemajor industrial risk areas of Sicily, and further ex-tended these networks with new network manage-ment systems.

9.6 SOFTWARE PACKAGE FOR CIVIL PROTECTIONEMERGENCY PLANNING & MANAGEMENTJ.P. Nordvik, M. Delaval

A software package (EPMS) has been developed tosupport the activities of civil protection authorities re-lated to planning and management of emergencies.Computer tools have been developed to support thecollection and diffusion of all civil protection relatedinformation, their organisation into ready-to-use emer-gency plans and procedures, and their exploitationduring real emergencies to support the different aspectsof the management of the crisis including in particularco-operation with all players involved in the crisis.

The product is sub-divided into two main software logicalunits: emergency planning and emergency managementapplications. The former is concerned with the prepa-ration of the emergency plans and procedures of civilprotection, with the possibility of geo-referencing thisinformation on the territory. It also concerns the manage-ment, updating, sharing and diffusion of this informationand of the emergency procedures concerning local,provincial or regional end-users. The latter offers to theoperator of an emergency room a software environmentto support the management of the crisis. Pre-definedprocedures of intervention are available for immediateuse by the operator. A “logbook feature” is availableto trace all actions performed by the operator for post-emergency analysis or emergency training procedures.The product will be validated through a realistic testcase involving all the players concerned during anemergency, i.e., local, provincial, regional administra-tions and operative structures of the civil protection

and other technical structures. Data, modules (forms),procedures and alarms are the most important entitiesthat will concern the users.

The Oracle server 8.0 and a first prototype versionof the EPMS software have been installed at ISIS fordemonstration and for a first assessment. A typicaluser-interface interaction is shown in Figure 9.4.

9.7 SAFETY OF INDUSTRIAL PROCESSESAND HUMAN FACTORSP.C. Cacciabuehttp://humanfactors.jrc.it/EAM2000.html

A research activity, performed in collaboration withthe Politecnico di Milano, focused on the importance ofcombining macro and micro approaches in the analysisof accidents and incidents in complex systems such asthose involving production and distribution of energy. Anew formalism was proposed, providing a working framefor showing the complex interconnections between ex-ternal and internal causes leading to the manifestationof erroneous behaviours [9.18].

The impact of human factors and in particular thequestion of safety-related effects have been investigat-ed and are recognised as the most important issues forthe future development of technology. Improving thesafety of systems implies improving the design, imple-mentation, evaluation and management of the associat-ed human factors and human-machine interfaces. Inseveral technological domains, the recognition of thiscrucial issue and the outcome of related research inhuman factors, has already led to the introduction ofstandardised methods in the design and developmentof automated systems, in safety assessment and intraining [9.19-9.20].

In other domains, the contribution of human factors tothe development of systems and working environmentsstill remains underestimated. For these reasons, thehuman factors issue needs to be further considered byregulatory bodies and developed at the technical level.

9.4 Typical forms displayed during an EPMS interaction.

Moreover, as this subject spreads throughout the wholeof Europe (and the world), the question of safety becomesa trans-national one, demanding the use of internation-ally accepted means and methods. In this respect, ISIShas organised and hosted the XIX European Conferenceon Human Decision-Making and Manual Control [9.21]and [http://humanfactors.jrc.it/EAM2000.html/].

9.8 THEMATIC NETWORK ON SAFETY ASSESSMENTIN WATERBORNE TRANSPORT (THEMES)P.C. Cacciabue

The objective of THEMES is to improve shipping safetyand marine environmental protection by establishinga common knowledge base and a framework of safetyassessment/management for waterborne transport.

This is met by:• describing the state-of-the-art for aspects related to

safety assessment including human factor issues;• establishing a common framework for safety assess-

ment based on present standards;• describing requirements and implementation strat-

egies for improved knowledge on risk factors;• identifying knowledge gaps and proposing further

research to be carried out;• making available and disseminating present knowl-

edge for safety assessments.

This network is co-ordinated by Det Norske Veritasand about 30 organisations, ranging from industry toacademia, are involved.

9.9 AUTOMOTIVE COLLISION AVOIDANCE SYSTEM P.C. Cacciabue, D. PetracconeThe activity, performed in collaboration with Politecnicodi Milano [9.22], focused on the development of a proto-type for a collision avoidance system (CAS) to be in-stalled in a car, according to the human-centred designphilosophy. Particular emphasis was put on the driver’sinterface. In this study the leading idea was also totake advantage of the know-how gained in the aero-space domain and see whether it is possible, and towhich extent, transfer the principles of human-centredautomation from aviation to automotive domains. Thesystem architecture was sketched together with thedesign of a head-up display (HUD). A demonstrator forthe HUD has been developed to show the main featuresof the interface, the automation issues in different situ-ations, i.e., nominal and abnormal conditions (poorvisibility, mechanical failure, and wet road) – (Figure 9.5).


References

9.18 Citterio, S, Invernizzi F. – Perspectives of integrationfor cognitive ergonomics within the safety manage-ment of complex systems – M.Sc. Graduation Thesisin Management and Production Engineering(in Italian), Politecnico di Milano, Italy, 2000.

9.19 Cacciabue P. C. – Human Factors Impact on RiskAnalysis of Complex Systems – Journal of HazardousMaterials, 71, pp. 101-116, 2000.

9.20 Cacciabue P.C. (Ed.) – Principles of Human ErrorManagement (Principi di gestione dell’erroreumano) (in Italian) – In G. Mantovani (Ed.)Ergonomia: lavoro, sicurezza e nuove tecnologie.pp. 81-122. ISBN88-15-07597-6. Il Mulino, Bologna.

9.21 Cacciabue P.C., (Ed.) – Proceedings of 19th EuropeanAnnual Conference on Human Decision Making andManual Control (EAM-2000) – ISBN 92-828-9583-1.Ispra (VA), Italy, 26-28 June 2000, EUR 19599-EN.

References

9.18 Petraccone, D. – Human-Machine Interaction withinan Automotive Collision Avoidance System: A Studyfor a Graphical Interface – M.Sc. Graduation Thesis inAeronautical Engineering (in Italian), Politecnico diMilano, Italy, 2000.

9.5 Screen shots of the demonstrator of the CAS.

Sensorsint. environment

Sensorsext. environment

Elaboration &mgmt. of data

InterfaceDiagnosis

Sensorsint. environment

Sensorsext. environment

Elaboration &mgmt. of data

InterfaceDiagnosis

9.10 MONITORING OF LANDSLIDESD. Tarchi

Synthetic Aperture Radar Interferometry (D-InSAR) usingERS 1-2 and RADARSAT data has currently generatedmore and more interest in the scientific community. Inthe field of earth sciences, particular interest is relatedto the construction of precise digital elevation modelsand new applications such as displacement detectionof earthquakes, landslides and subsidence phenomenahave been approached and validated.

ISIS has developed a specific methodology for theapplication of such a technique to the monitoring oflandslides based on the use of ground based imagingradar instrumentation. This approach combines thewell-known advantages of radar imaging techniques,such as remote action, day and night operability, non-invasiveness, with a maximum flexibility in terms ofviewing geometry, frequency of observation, polari-metric capabilities and temporal coverage.

The instrument, called LISA (Linear Synthetic ApertureRadar), is capable of performing measurements on areasextending from a few metres to a few kilometres thusproviding also close-in views of portions of particularinterest (Figure 9.6). The frequency of observation canbe selected in the range from 500 MHz to 18 GHz withfull polarimetric capabilities. Attainable spatial resolu-tions are in the range from tenths of centimetres to afew metres while the accuracy in the measurement ofdisplacements is in the millimetre range. Different ver-sions of this instrument are available, which are suita-ble for medium and long term monitoring but whichcan also be operated in case of adverse environmentalconditions or during emergencies when flexibility andminimum time for installation are required.

During 2000, several experimental campaigns havebeen carried out with the system, aimed at validatingthe approach. Sites for the experiments have been sel-ected in order to be representative of typical environ-mental conditions and characteristic of the phenome-non being monitored. Figure 9.7 shows a picture of atest area – the Tessina landslides near Belluno, Italy.In Figures 9.8 and 9.9, a whole series are shown. Theyhave been obtained by comparing pairs of imagesacquired at time intervals of 12 and 24 minutes, respec-tively. Colours, coded according to the scale, representthe relative displacement along the line of sight of thesystem of the different portions of the area with respectto the reference acquisition. Negative displacementsindicate displacements towards the observer. As con-firmed by the network of traditional optical devicesinstalled on the site the area was undergoing a quiterapid and complex pattern of displacements due to in-tense rainfall in the previous days. In both interfero-

grams, two main areas of movements are recognisable.The left bottom side shows the most rapid displacement.After only 12 minutes the relative displacement exceedshalf of the wavelength of the radiation employed, i.e.,9 mm, producing a fringe in the resulting interfero-gram. This common effect, sometime misleading in theinterpretation of the results, can be easily fixed by app-lying appropriate unwrapping algorithms. Increasingdisplacements are recognisable in the second interfero-gram, corresponding to a longer time interval, whereadditional fringes are present. All the data acquiredduring the experimental campaign have been carefullyanalysed and compared to ground truth data comingfrom traditional monitoring devices. Correlation hasbeen found in all cases leading to a full validation of theproposed approach as a valid complement to the existingtools commonly employed in monitoring operations.

Further activities have been already scheduled. Two add-itional aspects will be approached. The first will investi-gate the potential of the combined use of ground basedSAR platforms and spaceborne imagery in the field ofregional scale geological risk assessment. The secondoriginates from the interest for the technique manifest-ed by different Civil Protection Authorities and will aim atvalidating the approach in real emergency situations.

64

9.6 Picture of the small version of the LISA system during an

experimental campaign.

9.7 The Tessina landslide

ISIS is also active in the field of innovative SAR sensorsand concept development and testing such as noiseradar technology and parasitic SAR system. The latterconcept in particular, due to the absence of electro-magnetic emissions, is suited for long term monitoringof subsidence phenomena in urban areas. An opera-tional prototype is almost ready and will be tested inreal conditions in 2001.

9.11 DETECTION OF OIL SPILLSD. Tarchi, P. Pavlakis

The MARPOL 73/78 convention, recognising the spe-cific oceanographic and ecological sensitivity of anumber of regional seas, identifies them as “SpecialSeas Areas”. Within such areas, discharges of oil oroily mixtures from vessels are virtually prohibited. Allthe European regional seas are accorded with such astatus. Key elements, however, for the success of anyinternational environmental agreement, are monitor-ing and intervention. In the case of deliberate pollu-tion, monitoring is effective only when supported byperpetual airborne surveillance. It is, however, neithertechnically nor economically possible to employ air-borne surveillance possibilities over all the Europeanseas. Thus, spaceborne surveillance means need to beexploited, especially SAR (Synthetic Aperture Radar)sensors, which are independent of sunlight or cloudcoverage and are capable of providing wide-area rec-onnaissance. As a support to the European Commis-sion’s effort to ensure an optimum level of protectionfor the European marine and coastal environment, ISISperforms systematic work around such a quest. Throughthe best available techniques and methodologies andan established expertise, ISIS provides comprehensiveinformation on vessel pollution at the regional level.


40

35

30

5 10 15 20 25

European UnionEU Candidate satate“Special Sea Area” statusunder MARPOL 73/78

13 14 15

37

36

9.9 Displacements pattern (with respect to a reference

condition) after a time interval of 24 minutes.

900

820

740

660

580

500

420

340

260

180

100

-400 -320 -240 -160 -80 0 80 160 240 320 400

metre

met

re

4.5

3.5

2.5

1.5

0.5

-0.5

-1.5

-2.5

-3.5

-4.5

disp

lace

men

t (nm

)

9.8 Displacements pattern (with respect to a reference

condition) after a time interval of 12 minutes.

900

820

740

660

580

500

420

340

260

180

100

-400 -320 -240 -160 -80 0 80 160 240 320 400

metre

met

re

4.5

3.5

2.5

1.5

0.5

-0.5

-1.5

-2.5

-3.5

-4.5

disp

lace

men

t (nm

)

9.10 Spills detected on ERS-1/2 SAR images during 1999 in the

Mediterranean Sea. Shape and radar backscattering

elements of the spills, stored together with auxiliary

environmental information, help to reconstruct the spilling

scenario and associate the spill with ship tracks detected in

the pollution zone.

OVERVIEWD. BainChanges in research priorities within ISIS meant that2000 was a year of transition within the Decision Supportactivity area. 1999 had already witnessed a change offocus from exclusive concern with environmental man-agement projects towards sustainable mobility. Thisexperience in transferring competencies now provesinvaluable in making the further transition to risk andemergency management decision support.

As in the past, ISIS derives its expertise in the decisionsupport area from the tools and techniques which ithas developed and from the cumulative experiencegained in building customised Decision Support Sys-tems (DSS) for specific real-life projects. Much effortduring 2000 was devoted to maintaining, upgradingand developing the various decision support tools andto ensuring continued capability in their use. Specificoutputs in the area of risk and emergency planning aretherefore planned for later years.

Outputs in 2000 included a number of case studiesinvolving use of Multi-Criteria Decision Aids (MCDA) inareas such as water resource management and solidurban waste management. In addition, a new softwaretool, known as MARPA (Multi-criteria Analysis usingthe Reference Point Approach) was developed as aweb-friendly addition to the existing tool box. Workalso continued on enhancing scientific support to agri-environmental policies (section 10.1).

Parallel to preparation for the new tasks, work hascontinued on a number of competitive actions. Althoughnone of these is specifically related to the risk andemergency management theme, the various decisionsupport tools developed or under development haveillustrated important lessons, which will prove invalu-able in future projects. The projects in question includetwo shared-cost actions – MATADOR and UTOPIA (bothnow completed) – in the area of sustainable mobilityand CommonGIS, which advances the use of spatialanalysis techniques. There is also an on-going thirdparty work project called MURST, which deals withflood protection.

10.1 ENHANCING SCIENTIFIC UNDERPINNINGTO AGRI-ENVIRONMENT POLICIESD.H.A. Al-Khudhairy, C. Leemhuis, V. Hoffmann,J.M. Terres, R. Calaon, I.M. Shepherdhttp://intelligence.jrc.cec.eu.int/

Agriculture is the biggest land user in the EuropeanUnion (occupying approximately 40 % of its territory).Over the centuries, European agriculture has givenrise to unique European landscapes with a rich varietyof semi-natural habitats and species. However, in thelast few decades European agriculture has also beenviewed as a major source of pollution and a drivingforce behind the disappearance and degradation ofEurope’s landscape and wetlands. This is becauseEuropean agriculture has changed notably in the lastfour decades. It has become a complex “high-tech”industry that has increased agricultural production,thanks to rapid application of new technology and theencouragement of generous support under the CommonAgricultural Policy (CAP), by rates that far exceedthose in other industries. Moreover, because of thehigh level of agricultural land in and around remainingEuropean ecosystems of special biodiversity and wild-life interest, their future is strongly linked with changesin agricultural practices and intensity, and thus inreforms in agri-environmental policies.

The extent of environmental problems linked withEuropean agriculture and the lack of information avail-able to meet the demand of agri-environmental policy-makers, the farmers, the scientific community, and thegeneral public to understand agri-environmental rela-tionships, and to analyse the environmental impactsof reforms in agri-environmental policies has resultedin the considerable global interest shown today indeveloping and incorporating appropriate indicatorsinto agri-environment and agricultural policies. As partof the remits of the SHYLOC project1 and the JRC agri-environment cluster2, computer-based models, remotesensing based software, and specific agri-environmentindicators have been developed to provide scientificunder-pinning to agri-environment and agriculturalpolicy development, monitoring and analysis.

The computer-based model is an advanced wetlandmodel that will enhance the design and implementa-tion of agri-environment programmes for maintaining,protecting, or restoring wildlife habitats in agriculturallandscapes [10.1]. The remote sensing based software,

Decision Support for Risk and Emergency Planning10

66

1 SHYLOC (System for HYdrology using Land Observation for model Calibration) is a pilot project that started in November 1997and ended in October 2000. The project was partly funded by the European Commission, led by ISIS-JRC, and had partners inEngland (University College London), Denmark (DHI Water & Environment) and Greece (University of Thessaloniki & GreekBiotope/Wetland Centre).

2 The agri-environment cluster comprises several JRC institutes including ISIS, SAI, EI and IPTS, and interfaces with the Environ-ment DG, the Health and Consumer Protection DG, the Energy and Transport DG, EUROSTAT and the European EnvironmentalAgency, through participating and providing support in the working group on “Indicators Integration in the CAP”.

SHYLOC [10.2], was developed by ISIS and validatedagainst ground-measurements provided by SHYLOCpartners to provide information on specific environ-mental pressure-state indicators by means of monitor-ing (a) surface water levels in areas of water manage-ment/abstraction issues where farmers receive pay-ments to maintain high water levels in ditches anddykes to maintain and enhance wildlife conservationvalues of wetlands and (b) water level requirements ofwetland flora and fauna in ditches for which farmersreceive payments to manage and maintain high ditchwater levels. Finally, the framework and methodolo-gies for structuring and analysing a set of specific agri-environmental indicators and related environmentallinks that the JRC can provide in the short, medium andlong term have been established, and are reported in[10.3].

The afore-mentioned models, remote-sensing basedsoftware, and agri-environment indicators form part ofthe mission of the JRC as a service of the EuropeanCommission to foster sound EU agri-environment andagricultural policies through credible scientific andtechnological under-pinning, and hence to contributetowards the long-term protection of the environment.


SHYLOC

Vector data(outlined in red)

Ditch carryingpixels(shaded in grey)

Wet ditch widthsGround measurements of

water levels

Ditch carrying pixelscovered in water

Statistical relationships between satellite-derivedresults and ground measurements

Ground control points

Geometrically corrected image

10.1 Integrating satellite-derived information with ground measurements provides scientific support to assessing the

environmental effects of EU policies, such as agriculture, on areas of special biodiversity/nature interest, and to

monitoring the conformity of the environmental objectives of such policies to environmental targets.

References

10.1 Al-Khudhairy, D.H.A., Leemhuis C., Hoffmann V.,

Shepherd I.M., Thompson J.R., Gavin H.,

Gasca Tucker D., Zalidis G., Bilas G.,

Refstrup Sørenson H., Refsgaard A., Papadimos D.,

Argentieri A. – SHYLOC Final Report – EUR 19755EN,European Commission (2001).

10.2Al-Khudhairy, D.H.A., Hoffmann V., Leemhuis C.

– SHYLOC User Manual – EUR 19745 EN, EuropeanCommission (2001).

10.3Terres J.M., Al-Khudhairy D.H.A. – Possible contribu-tions by JRC to meet the needs of the Commission onagri-environment indicators – EUR 19639 EN, Euro-pean Commission (2000).

68

10.2 MATADORD. Bain, F. Mazzeo Rinaldi, R. Wolfler Calvo,L. Trovelli

The use of non-conventional vehicle drive-line techno-logies (electric, gas, hybrid, etc.) is an increasingly liveoption for fleet operators in areas as diverse as car-hire, bus services and road haulage. Occasionally theinitiative for considering such options may come fromfleet managers themselves. More frequently, however,the pressure to examine such alternatives will comefrom other sources, including environment and healthlegislation, local access regulations, incentives offeredby state agencies or from other parts of the organisa-tion keen on improving the corporate image.

As fleet managers will readily testify, the introductionof new technologies is generally a highly complex opera-tion, with issues such as maintenance, reliability, range,refuelling structures, driver acceptance, staff training,vehicle behaviour, re-sale values and availability ofspares as important as the headline characteristics suchas fuel costs per kilometre and environmental impacts.Comparative information on all these matters may bedifficult to obtain.

The MATADOR project (Management Tool for the Ass-essment of Drive-line Technologies and Research) seeksto provide fleet operators (and others interested incommercial introduction of new propulsion techno-logies) with a wide range of information, much of it coll-ected for the first time. The software tool that guidesthis process is known as MIG (MATADOR InteractiveGuide). As its name implies, MIG help users to accessthe MATADOR information base and will also givethem the possibility of making a multi-criteria compar-ison of various new drive-line options, relative not onlyto each other but also to conventional vehicles.

Exploring the details of the different vehicle typesavailable allows the user to appreciate both positiveand negative attributes and the impact of these on arange of different decision criteria. While single criteriamay be sufficient to discard particular options it isextremely rare that they can form the basis of accept-ance. MIG incorporates a decision enhancement capa-bility based on Multi-Criteria Analysis (MCA) to assistfleet operators to take a range of vehicle characteristicsinto account before making any decision to proceed.

The software structure of the MIG has been designedin terms of software modules, module interactions,functionalities, general information contents and rulesfor the delivery of inputs in terms of models, data anddocumentation.

10.3 UTOPIAD. Bain, F. Mazzeo Rinaldi, R. Wolfler Calvo,L. Trovellihttp://utopia.jrc.it

Most Europeans live in cities and even most of thosewho do not are to some degree dependent on cities foremployment, education, entertainment and otherservices. How cities work, the quality and dynamism ofthe urban system are therefore of vital importance tothe vast majority of European citizens. While it is, ofcourse, people who are the key element in determin-ing the relative success of cities, physical aspects havea crucial shaping influence. Most elements of this –housing, sanitation, lighting, shops, offices, etc. – haveimproved dramatically over the last century. The onearea where there has been a marked deterioration, fora large proportion of European cities, has been intransport.

This has come about through an explosion of mechani-cally-assisted mobility, particularly the vast growth inthe numbers of private cars entering and using the urbanspace. The problems that these raise have led to wide-spread calls for radical change. Among the solutionsproposed are physical or financial disincentives to usecars, encouragement of other travel modes such aspublic transport or cycling and walking, promotion ofvirtual mobility using advanced information techno-logy, and the adoption of new propulsion systems.

UTOPIA focuses on this last option. It considers thewide range of new and improved propulsion systemscurrently coming on stream and draws together findingsfrom a wide range of disciplines to address the complexissue of how to ensure successful market entry of newtransport technologies. Although the main driver ofthe introduction of such technologies is concern forthe environment they will also have a wide range ofimpacts on many other aspects of society. These addi-tional impacts need to be understood and given dueweight by decision-makers – apart from their intrinsic

10.2The Utopia homepage.


importance they also affect the likelihood of markettake-up. UTOPIA draws on practical experience derivedfrom previous case studies in Europe, giving guidanceon how these can best be applied to specific urbansituations and providing tools to understand what theeffects might be.

NAVIGATE UTOPIA is a software tool which has beendeveloped to allow local decision-makers to exploreand evaluate the numerous new transport technologyoptions already available or on the point of enteringthe market. Getting urban transport “right” is a chal-lenging task for decision-makers, given the number ofpolicy areas affected, the large range of stake-holdersand the high political sensitivity of almost any optionadopted (including “do nothing”). Ultimately any deci-sion must rest with the skills and informed judgementof the decision-maker or decision college. It remainsdifficult. NAVIGATE UTOPIA can help, however, by pro-viding structured access to a large quantity of informa-tion and experience, allowing systematic balancingof a wide range of options and consequences andallowing serious “what if?” speculation.

The system is meant to assist decision-makers in ex-ploring and assessing trade-offs between complexquantitative and qualitative information in an integrat-ed and balanced way, applying their own value judge-ments.

10.4 THE COMMONGIS PROJECTR.J. Peckham, S. Maffulli

ISIS continued its participation in the CommonGIS, ashared-cost action project funded by the ESPRIT pro-gramme. The aim is to develop the means for the dis-semination of geo-referenced information to a broadcross section of the public via the World Wide Web. Itdelivers map-based visualisations of statistical datathrough an Internet browser, together with user friendlymethods for interactively adjusting the displays. TheInternet is increasingly being used as a means of dis-seminating information not only among specialistsand professionals, but also for the general public. Inmany areas of decision-making presentation of the in-formation on maps is a powerful aid to perception ofthe phenomena concerned and is essential to under-standing spatially distributed phenomena.

The project has proceeded through the developmentof a first prototype, user studies and evaluations, andthen a second prototype incorporating the feedbackand suggestions from the trial users. The project con-sortium included partners from Germany, the Nether-lands, Portugal and Italy. ISIS contributes to the projectmanagement, to application development, and infor-mation management, as well as standardisation anddissemination activities.

The system developed includes facilities for automaticgeneration of visualisation instruments such as colour-shaded areas, bar charts and pie charts superimposedon maps within a web browser. Intelligent, knowledgebased methods are used to automate the generationof visualisations and to minimize the inputs requiredfrom the user. Once they have been generated the visual-isations can be interactively adjusted using familiarcontrols to aid exploration and understanding the data.

The system has many potential applications and during2000, ISIS has set up two prototype applications onforest fires and earthquakes. The forest fires demon-strator shows on a map the mean duration of fires, themean delay between the alarm and the interventionand the total surface burned, referred to the provincesof Spain. In the earthquake application, data from aEuropean catalogue of historical earthquakes are usedto illustrate the basic attributes of the events (magni-tude, intensity and radius) using symbols with gradu-ated colours. The user can interactively adjust a refer-ence value above and below which the symbols aredisplayed in different colours. Alternatively a singleevent can be chosen as the reference and events withparameter values higher and lower than this can berapidly visualized. The events can also be displayed asbars with variable colours and heights, again with userfriendly tools for adjusting the visualisation.

NAVIGATE UTOPIA andthe UTOPIA Decision Framework

Objective: To provide structured decision support on appropriate transport solutions (including technology choice) and the need for pilot

and demonstration projects, linked toinformation resources covering available options and case study experience.

Scope: Pre-screening of options, to help decision-makers home in on a particular course of action. (The preferred solution would then need a more detailed assessment, and anypilot or demonstration element can be sup-ported by the UTOPIA Project Guidelines.)

Target user Transport planners, policy-makers andgroup: fleet operators in medium and large cities

who perceive that they may need to intro-duce initiatives based on new transport concepts and cleaner propulsion systems in order to tackle local problems.

Product: A web-style tool based on Multi CriteriaDecision Aid for defining and comparing options, linked to guidance on issues to consider, and case study material.

70

Tools of this kind can be effective aids to decision mak-ing, and when implemented over the Internet they canfacilitate access to, and understanding of, geo-refe-renced statistical information by a broad spread of in-terested stakeholders and managers. However it is asignificant challenge to make these tools sufficientlyuser friendly and intuitive to be used by non-specialists.The user studies carried in the CommonGIS projecthave helped to make progress in this area.

10.5 THE MURST PROJECTF. Mazzeo Rinaldi, R. Wolfler Calvo

The project is a third-party work financed by the Minis-

tero dell’Università e della Ricerca Scientifica e Tecno-

logica (MURST). The task of ISIS is to develop a multi-criteria decision support system for flood protection.The application addresses flood damage prevention andcomprises a Geographical Information System (GIS)together with three models to simulate the behaviourof the rivers and streams in flood (a distributed hydro-logical model, a semi-distributed model and a model offlood propagation along riverbeds) and a multi-criteriamodule.

The goal of the decision support system, called DEFF(Decision Framework for Flood Protection), is to supportdecision-makers in finding a compromise solutionamong a set of possible actions aimed to achieve anoptimal use of the territory, the safety of people and areduction of the damages caused by critical floodevents. DEFF will be applied to a case study in the LiriGarigliano watersheds. The Liri Garigliano watershedis one of the most important in the south of Italy andextends over the territory of three regions (Abruzzi,Lazio and Campania) and five provinces (L’Aquila,Frosinone, Latina, Roma, Caserta). It takes water froma large section of the south-western Appennines anddischarges it in the Tirreno sea.

The basin’s environmental system, as produced bycenturies of close interaction between natural evolu-tion and human-made changes, now appears to a largeextent compromised. The strong land modification byinfrastructure of the last 30 years, coupled with thenatural vulnerability of the basin landscape, has causeddamage, some of which is irreversible. Basin authoritieshave initiated preliminary studies to better plan futureactions to protect the Liri Garigliano watershed. Badland management is the main cause: on the one sidecongested urban and industrial areas overcharge riverswith polluting discharges and, on the other, in marginaland abandoned rural areas the loss of population andof traditional activity makes lands vulnerable to erosionand destabilisation.

A publication of the Liri Garigliano basin authority high-lights the following emergencies and critical points:• pollution and overuse of waters, with critical values at

Fucino, Sacco, Melfa and at the mouth of Garigliano;• destabilisation of slopes;• flooding, with particularly high risk in the middle Liri

valley and Sora plain, along the greatest part of theSacco stream and in the low Garigliano area;

• land degradation, with critical areas in the high Saccovalley, at the Cosa inflow in Sacco and in Liri Valleybetween Sacco and Melfa down to Cassino.

10.3 A screen image from the first prototype of CommonGIS.

The system provides visualisation of statistical data on

maps via an ordinary Internet browser. Here population

data for Portugal are being examined as colour shaded

areas.

Region and provinces of the watershed.

Data from Liri Garigliano and Volturno basin authority.

Region Province Surface Surface falling in

the watershed

Abruzzo L'Aquila 5034 km2 1242 km2

Lazio Frosinone 3239 km2 2825 km2

Lazio Latina 2251 km2 296 km2

Lazio Roma 5352 km2 474 km2

Campania Caserta 2639 km2 287 km2

10.4 Watershed view.

71


Brand C., Mattarelli M., Moon D., Wolfler Calvo R.

– A Decision Support System for the operationalmanagement of water-supply networks – EuropeanJournal of Operational Research (forthcoming).

Brand C., Mattarelli M., Moon D., Wolfler Calvo R.

– STEEDS: a Strategic Transport/Energy/EnvironmentDecision Support – European Journal of OperationalResearch (forthcoming).

Florio L., Mussone L, Wolfler Calvo R.

– A model to calculate the capacity of complex railwaysystem – Transportation Research part. B (forthcoming).

Cordone R., Wolfler Calvo R. – A heuristic for the VehicleRouting Problem with Time Windows – Journal ofHeuristics (forthcoming).

Linde N., Eliasson Å, Mazzeo Rinaldi F.

– Multi-Criteria Decision Aid as a Tool in WaterManagement – Technical Note No.I.00.143,December 2000.

Angioni S., Cinus S., Mazzeo Rinaldi F., Wolfler Calvo R.

– A Decision Support System for InterventionsPlanning Aimed at Flood Damage Prevention –in Proceeding of European Conference on Advances inFlood Research - Potsdam 1-3 November 2000.

Mattarelli M., Moon D., Wolfler Calvo R.

– STEEDS: a Strategic Transport/Energy/Environment Decision Support – in Proceedings ofHCP’99 10th Mini EURO Conference, pp 377 – 382,Brest, 22 – 24 September 2000.

Gadda A., Mazzeo Rinaldi F., Wolfler Calvo R.

– SOCRATE - SOCietal Responses Analysis toEnvironmenTal issuE – in Proceedings of HCP’99 10th

Mini EURO Conference, pp. 345 – 352, Brest, 22 – 24September 2000.

Cavallet A., Malucelli F., Wolfler Calvo R.

– A non-linear multi-commodity network designapproach to solve a location-allocation problem infreight transportation – ODYSSEUS 2000 Workshopon freight transportation and logistics, Chania, May2000.


In order to improve the present situation, several alter-natives can be proposed. In the choice of a better solu-tion major help is given by multi-criteria methodologies:alternatives are compared by means of criteria thatcan be satisfied at various levels. This case of a decisionsupport system for mitigation and prevention of dam-ages due to floods comprises four families of criteria:environmental, economical, social and technological.Each alternative involves an integration of structuralinterventions that tend to reduce the probability of thecritical event and non-structural interventions such asregulation in the possible use of a certain area, the or-ganisation of monitoring of rivers and assistance tothe people during emergencies. Risk = 0 is impossibleto achieve, hence it is necessary to limit the comparisonof alternatives amongst those which can reduce therisk to an acceptable level.

11

72

Aviation Safety

OVERVIEWA.C. LuciaThe projects presented in this chapter are part of theJRC effort supporting the implementation of policiesaimed at the protection of individuals, infrastructuresand society against natural and technological hazards.The core activity in aviation safety is constituted by theinstitutional project on the European Co-ordination Cen-tre for Aviation Incident Reporting Systems (ECCAIRS),supporting the Energy and Transport DG in the estab-lishment of a common European reporting system forin-flight incidents, including analysis of the influenceof human error. The Energy and Transport DG has re-cently prepared a proposal for a Council Directive“Establishing a co-ordinated system of national man-datory occurrence reporting schemes”. The proposalis based on the conviction that the establishment atthe Community level of a mandatory aviation incidentreporting system is feasible and that a suitable tool(ECCAIRS) is available to those Member States that donot want to develop their own system. A centraliseddatabase facilitates the exchange of information be-tween all Member States.

Several synergistic competitive activities are also on-going, focusing mainly on human factors, which con-tribute to keeping and updating ISIS’s expertise in thisfield. Based on the results obtained in 1999 and 2000,the priority activity development lines for the futurehave been identified. They include: maintaining, up-dating and further developing ECCAIRS as well as pro-viding support to the implementation of the relevantDirective; supporting pre-accession countries and facili-tating their gradual involvement in ECCAIRS; strength-ening data analysis activities by a co-ordination effortat the European level; and starting a risk analysisactivity on some priority aspects (at present under dis-cussion with the Energy and Transport DG) of aviationsafety.

11.1 EUROPEAN CO-ORDINATION CENTRE FORAVIATION INCIDENT REPORTING SYSTEMS(ECCAIRS)

W. Posthttp://eccairs-www.jrc.it/

All safety experts recognise that the global rate of avia-tion accidents is stabilising and, as a consequence, thegrowth in air traffic, if nothing is done to improve it, willlead to an increase in the absolute number of incidentsand accidents per year. For that reason the EuropeanCommission (the Energy and Transport DG) proposedto create the necessary framework to collect and dis-seminate information on aviation incidents, at the widestscale, so that all parties involved in aviation can learnfrom mishaps and improve their performances to producea safer system.

The related directive1 “Proposal for a directive of theEuropean Parliament and of the Council on occurrencereporting in civil aviation”, prepared by the Energy andTransport DG in December 2000, will address bothmandatory and confidential reporting of incidents, de-fects or malfunctions, which may constitute a hazardfor civil aviation operations (called under the genericterm of “occurrences”). Since accidents are usuallypreceded by a number of these “precursors”, a betterknowledge of occurrences may help to eliminate someof the causal factors leading to an accident and there-fore prevent it.

The ECCAIRS project provides support to the Energyand Transport DG for the setting-up and implementa-tion of the framework described. ECCAIRS is a networkof collaborating EU aviation authorities, set-up by JRC-ISIS over the last 5 years, whose objective is to ex-change, collect and assess aviation incident and acci-dent information provided by EU aviation authorities.The project’s objective is to have all Member Statesparticipa-ting actively in the network by the end of theFifth Framework Programme (2002) when, most likely,the European Commission’s directive will have beenadopted.

In 2000, the ECCAIRS steering committee, in whichrepresentatives of all EU Member States and otherslike the Energy and Transport DG, ICAO2 and Eurocontrol3

have a seat, had its second annual meeting in Ispra.At the meeting the steering committee decided to im-plement in the year 2000 an enhanced version of theECCAIRS 3.3 reporting system, a tool already widelyused by many countries (Ireland, Denmark, France,Greece, Germany and Italy) and under test or evaluationby most other EU aviation authorities. In December2000, the development of this new version, Release3.4 has been completed. The French Civil Aviation Au-thority (CAA) developed, using tools provided by theJRC, a French version of the ECCAIRS application forthis release, which is now being used locally. The datacollected, apart from narratives, remains completelycompatible with the English version and can be ex-changed without problems. The German AIB has pro-totyped a similar German version to be completed inthe first months of 2001.

Also during 2000 the preparation and prototype develop-ment of a completely new, but backwards compatible

1 Document COM(2000) 847 final 2000/0343 (COD) of

19.12.20002 International Civil Aviation Organisation, part of the

United Nations.3 The pan-European organisation responsible for

co-ordinating Air Traffic Management.


version of the reporting system (Release 4, see Figure

11.1) has been started and resulted in a second beta-version made available to the authorities in September.A very close collaboration with ICAO, Eurocontrol,Deutsche Flug Sicherung (DFS) and the Swedish CivilAviation Authority finally resulted in a proposal for anew data classification scheme to be used in ECCAIRS.The scheme is also highly compatible with Eurocontrol’sstandardisation efforts for Air Traffic Management (ATM)related occurrences. The result is that there is an increas-ing chance that the ECCAIRS reporting system will notonly be used by EU civil aviation authorities but also byorganisations at an international level (ICAO and non-EUauthorities) or in related areas like ATM (Eurocontroland DFS) and others.

11.2 HUMAN FACTORS IN ACCIDENT DATABASES P.C. CacciabueFollowing the research actions oriented towards accidentanalysis, safety assessment and human behaviourstudies, ISIS has been involved in the development ofthe new Accident Classification scheme (ADREP-2000)proposed by ICAO. In order to assess advantages andopen issues associated with ADREP-2000, especiallyin relation to human factors, an extended study wasstarted in 1999 and continued throughout the year2000 within the project ECCAIRS. This study has shownthe ability of ADREP-2000 in representing, in a con-densed and formal way, the findings of an accident,including cognitive and organisational aspects. In ad-dition, the new classification scheme was shown toincorporate the most modern theories of human factorsand human errors. However, the classification processwith ADREP-2000 remains quite complex and cumber-

some to complete, especially for the part related tohuman factors, and the risk of generating inappropriatecoding remains quite high. For this reason it is necessaryto develop a support tool that provides appropriateguidance in the application of the taxonomy.

A number of other considerations has resulted fromthe study [11.1]. First, considering that accidents in thedomain of aviation are very rare, the contribution ofthe insight contained in mandatory reporting databasesis too scarce and dispersed to offer a real contributionin terms of information and data. This implies that froma database on accidents it is not possible to identify realindicators of incipient failures of defences or conditionsof latent errors. Other data collection systems, e.g.,voluntary reporting systems on near-misses and inci-dents, are much more appropriate for this purpose andricher in information content as long as the collectedinformation is properly analysed.

Secondly, for the same reason as above, i.e., the scarc-ity of data, a database only focused on mandatory typeof events, such as ADREP, does not offer a sufficientamount of data to derive probabilities or failure ratesfor risk assessment and risk management methods[11.2]. As an example, Figure 11.2 shows a comparisonbetween the old ADREP-87 classification and ADREP-2000 based on 6671 accidents codified with the twotaxonomies. It is quite obvious how the probabilitiesthat can be associated to each type of error are onlybroadly evaluated by making substantial assumptionsabout behavioural and cognitive functions. Moreover,these data are dependent probabilities on the occurrenceand therefore have a very limited range of generality.

11.1 The new release 4 of the ECCAIRS reporting system.

In conclusion, although ADREP-2000 is able to describethe accident from an organisational and socio-technicalpoint of view, it is largely insufficient in offering an in-depth insight and is useless in promoting importantchanges and improvements. It does not offer a solidbase for the definition of “safety indicators” within anorganisation, and is not able to represent a consolidatedset of data for risk and reliability assessment. Only bycoupling accident data with findings from other sourcesof investigation and information such as voluntaryreporting, task analysis and regular assessments ofwork and safety practices is it possible to obtain realand effective insights into human factors for improvingsafety levels and preventing future occurrences.

A complementary activity, in the framework of thecommon research programme with the Politecnico diTorino (Aerospace Department), focuses on humanreliability methods, accident investigation methodsand related data collection and analysis. The aim ofthis research in the domain of aviation is to supportthe evaluation of the new classification schemes beingdeveloped within the ICAO. In the year 2000, the tax-onomy and classification system was analysed withrespect to its ability to generate human reliability datafor probabilistic safety studies. The most relevant con-clusion of this research was that, while ADREP-2000level of detail is fine and permits the reconstruction ofan accident for many aspects, especially related tohuman interactions, its ability to support reliabilitydata definition is presently questionable [11.3]. 11.3 COMPETITIVE ACTIVITIES ON

HUMAN FACTORS IN AVIATION SAFETYP.C. Cacciabue, M. Pedrali, M. Bacchi,D. Baranzini, A. Trasi

11.3.1 Aircraft Dispatch and Maintenance Safety

(ADAMS)

The project ADAMS aimed at the study and design ofan integrated human factors safety system for aircraftmaintenance and dispatch. The following partners tookpart in the project: Defence Evaluation and ResearchAgency (United Kingdom), Netherlands National Aero-space Laboratory, Trinity College Dublin (Ireland),Sabena (Belgium), Scandinavian Airlines System, AerLingus-TEAM and the Joint Research Centre (ISIS).

The project ended in early 2000. A “Human Factor Guide”was finally published by the consortium titled “HumanCentred Management for Aircraft Maintenance”. Thisguide is devoted to the industry and contains user-focused practices for managing human factors in aviationmaintenance. The guide (on CD-ROM) and the ADAMSoutputs were presented to industries and EuropeanAviation institutions during a dissemination day entitled“Human Factors: Beyond JAR Compliance – creatingeffectiveness in Aircraft Maintenance”, held in Amster-dam on the 28th February, 2000.

74

References

11.1 Cacciabue P. C. – Human Factors Insight andReliability Data from Accident Reports: The case ofADREP-2000 for Aviation Safety Assessment –Proceedings of PSAM 5 - International Conference onProbabilistic Safety Assessment and Management.Osaka, 27 November - 1 December, 2000.

11.2 Cacciabue P. C.

– Human Factors Insight and Data from AccidentReports: The case of ADREP-2000 for Aviation SafetyAssessment – 3rd International Conference onEngineering Psychology & Cognitive Ergonomics -EP&CE-III – Edinburgh, 25-27 October, 2000.

11.3 Piccoli, A. – Human factors in aviation accidents:classifications, data and safety analysis –M.Sc. Graduation Thesis in Aeronautical Engineering(in Italian), Politecnico di Torino, Italy, 2000.

11.2 ADREP-2000 human reliability data from 6671 accidents

codified with ADREP-87.

ADREP 2000 p ADREP 87

1 liveware - -

11 physical - -

111 physical characteristics - -

112 sensory limitations - -

113 other physical limitations

or characteristics. - -

12 physiological 2.7e-3 physiological factors

121 illness/incapacitation 3.9e-3 other physiological factor

122 health/fitness/ lifestyle 3.9e-3 other physiological factor

123 fatigue/alertness 2.7e-3 other fatigue

124 illusions 3.9e-5 other physiological factor

125 other physiological. limitations 3.9e-3 other physiological factor

13 psychological 8.99e-4 psychological condition

131 action/error 5.68e-2 action

132 planning 8.74e-2 planning

133 inform. proces./decision making 8.99e-4 psychological condition

134 attention perception

& monitoring 8.99e-4 psychological condition

135 skill/technique/ability 6.52e-2 airmenship

136 judgement 8.99e-4 psychological condition

137 knowledge acquisition/ learning 8.99e-4 psychological condition

138 situational awareness 8.99e-4 psychological condition

139 personality and attitudes 2.83e-2 self confidence

1310 mental/ emotional state 8.99e-4 psychological condition

14 workload management - -

141 prioritisation/ task shedding - -

142 timing - -

143 task shedding - -

144 task allocation - -

145 other workload management - -

15 experience, qualifications

& knowledge 3.48e-2 exp. on aircraft type

151 experience & qualifications 3.48e-2 exp. on aircraft type

152 recency 0 recency/other

153 knowledge 0 knowledge

Code


11.3.2 Development of a Method for Air Transport

Safety Improvement through Quantitative

Risk Evaluation (DESIRE)

The 3-year project DESIRE was completed by the endof 2000. The following partners took part in the project:National Aerospace Laboratory NLR (the Netherlands),Joint Research Centre (ISIS), Airbus Industrie, SextantAvionique (France), Marsh (UK) and Airclaims (UK).

The objective of the project was to develop and evaluatea quantitative risk assessment model of air transportsafety, which allows the cost-benefit analysis of safetymeasures. The project started with an assessment ofthe factors that govern the need to improve aviationsafety. New pressures on aviation safety, such as air-port congestion, declining financial resources, ageingaircraft, insufficient safety oversight in some parts ofthe world were identified. Safety assessment tech-niques that are currently being used were evaluatedand developments in risk-based regulation were re-viewed. The role of cost/benefit criteria used by deci-sion-makers in the aviation system was discussed in theperspective of current safety assessment techniques.

It was believed that by placing accidents into a limitednumber of broad groups, based on the sequence ofevents and the likely severity of their outcome in termsof physical damage and deaths and injuries, a modelcould be derived allowing predictions for “typical”groups of accidents. Hence, the model architecturewas centred on the “single-consequence” concept, atthe end of a chain of events. The consequence is de-fined as “the event in the accident sequence that result-ed in the most damage and/or deaths and injuries”.Detailed analyses of the expected outcome in terms ofaircraft damage and the death of the occupants werecarried out for each of the defined consequences.

A classification of causal factors in aviation accidentsand incidents was developed, which enabled the de-scription of the chain of events that can lead to a par-ticular consequence. The classification provides a setof categories for the creation of an accident scenario,in which both the events (i.e., factual information – the“whats”) and the causal factors (i.e., causal informa-tion – the ‘whys”) related to a consequence are high-lighted. For each of the categories, an attempt wasmade to quantify their frequency of occurrence in normaloperations [11.4].

In order to allow a cost benefit analysis, a classifica-tion of cost factors was also developed. Both directcost factors, such as those related to passenger deathsand injuries and aircraft physical damage, as well asindirect cost factors, such as airline loss of reputation,were considered. Financial data for a large sample of

some 1,000 accidents were reviewed in order to pro-duce a cost profile for each “consequence”. The cost-side of the model is not intended to predict the costs ofindividual accidents but rather to provide an indicationof broad average costs, which may arise from accidentsfalling into typical groups.

Although considerable care was taken in selecting thefactors, the interconnections, the probabilities and thefinancial equations and data, it is expected that userswill make their own choices to tailor the model to theirspecific situation. The model describes “average” acci-dents, both on the causal side as well as on the costside. In reality, each individual accident may have itsown peculiarities. The model should not be used tojustify cuts in the safety budget. It should be used asone of the many tools that can help to determine howto spend the safety budget in the most efficient manner.

11.3.3 Advanced Cross-Communication Environment

providing Support Services to Dispersed

Maintenance and Technical Support Engineers

(ACCESS-maints)

http://www.metriqs.it/accessmaints/index.htm/

The main objective of the project, started in February2000, is the implementation of a Corporate Knowledge(CK) Management and Delivery Platform that will permitexperienced technicians and young professionals toreceive appropriate support for their engineering ac-tivity. A second objective is the definition of methodol-ogies, rules and tools that will permit the collectionand the sharing of those parts of CK having direct im-pacts on the specific industrial areas. The CK does notbelong to one company but it is usually distributedthroughout the full chain provider-integrator-client.The following partners are taking part in the project:SIA-Società Italiana Avionica (Italy), MQS (Italy), MPL(Italy), Politecnico di Torino (Italy), CISI (Italy), CS(France), CASAS (Spain), GTD (Spain), ISPASAT (Spain)and the Joint Research Centre (ISIS).

The IT basis of the ACCESS project consists of the im-plementation of a communication and server platformenabling transfer of corporate knowledge between re-mote sites: the ACCESS Server Platform (ASP) and net-work infrastructures, that range from public switched

References

11.4 Roelen, A.C.L., Pedrali, M., Hayes, P., Mariton, T.L.

– The development of an aviation risk assessmentmodel which allows cost benefit analysis of safetymeasure – presented at PSAM 5, InternationalConference on Probabilistic Safety Assessment andManagement, Osaka, Japan, 27th November –1st December (2000).

76

Telecom networks to Internet and private networks usingif possible high speed satellite links. The ASP includesa database containing CK, which may be updated underthe control of experts from a given company.

The standpoint of the project is the definition of CK as“the set of values, beliefs, assumptions, symbols andbehaviour that identify an organisation in relation withother organisations and govern the ways in which rela-tionships, business and practices are carried out”. Theproject started with the identification of the real needsof users to manage, represent, store, transmit andshare parts of the CK and on the practices adopted toaccess this knowledge.

In the first year of work, the technical aspects of theASP, i.e., the IT structure on which the knowledge isstored and through which it is transferred and the basicmechanisms of CK transfer have been studied. In par-ticular the key figures in knowledge management, i.e.,the knowledge holder, knowledge broker and know-ledge seeker, have been identified and correlated withthe ASP structure.

11.3.4 Advanced Integrated Training

in Aeronautics Maintenance (AITRAM)

http://www.aitram.de/

The project aims to contribute to the improvement ofthe learning process by developing an advanced train-ing system for aeronautical maintenance technicians.This training will address both technical and humanfactors issues and will be based on innovative concepts,new cognitive approaches and simulation technologiessuch as Virtual Reality. The project started in May 2000.The following partners are participating in the project:Fraunhofer GesellschaftInstitut Fabrikbetrieb und Au-tomatisierung (Germany), Trinity College Dublin - TCD(Ireland), SR Technics (Switzerland), FLS Aerospace(Ireland), AirEurope (Italy) and Joint Research Centre(ISIS).

During 2000, ISIS was involved in the collection of userneeds through on the job observations and the appli-cation of the Aircraft Maintenance Attitudes Surveyquestionnaire, developed within the ADAMS projectand adapted for AITRAM purposes. Afterwards, ISISdeveloped an integrated training concept in collabora-tion with TCD, based on user needs assessment, thelatest results of the ADAMS project, and the state-of-the-art of training means (STAMINA course). Internalto the AITRAM project, the collaboration with TCD hasbeen established in order to develop new models ofaviation maintenance training systems focusing on aprogressive integration of human factors with technicaltraining elements by the use of Virtual Environmentdevices. A survey has also been conducted across the

three end-users of the AITRAM project to identify humanfactors and technical needs to be considered for imple-mentation. The use of a new version of the AircraftMaintenance Attitudes Survey served this purpose[11.5].

From this perspective, a preliminary integration oftechnical and human factor (HF) training elements hasbeen centred upon the development of a new compe-tence model for the Aircraft Maintenance technician.This model will integrate HF and technical competencyrequirements in order to satisfy those HF and technicaltraining objectives, which are most frequently appliedas separate elements in aviation maintenance domain.The AITRAM training system approach will be illustratedthrough the analysis of three different maintenancescenarios: aileron servo-control removal/installation,integrated drive generator removal/installation andnose landing gear seals removal/installation.

11.3.5 Aviation Safety Targets for Effective Regulation

(ASTER)

The ASTER consortium brings together: National Aero-space Laboratory – NLR and the Netherlands EconomicInstitute (the Netherlands), Meridiana (Italy) Airclaims(UK), Israel Aircraft Industries (Israel), Eurocontrol(France) and the Joint Research Centre (ISIS). The mainobjective is the development of a means to enablesafety targets to be set and optimised for each of theparticipants in the air transport system (ATS) in orderto reach the optimum level of safety for the system asa whole. To achieve that it is necessary to identifythe elements within the ATS (i.e., airports, air trafficmanagement, airworthiness of “products” and opera-tions) and the causal factors that might jeopardise thesafety of each element. Afterwards, a target level ofsafety (TLS) of the ATS can be established with whichto compare the current one. Through the collectionand analysis of data on occurrences it will be possibleto quantify the safety per element and check consist-ency against this TLS, leading to the identification ofthe best safety practice or measure to be adopted to“fill this gap”.

A classification, derived from the DESIRE project, sup-ports the identification of causal factors. A functionalmodel of the ATS, derived from a NLR study for the

References

11.5 Baranzini D., Bacchi M., Cacciabue P.C. – A Training,Evaluation and Research Tool for AircraftMaintenance Teams – 3rd International Conferenceon Engineering Psychology and CognitiveErgonomics - EP&CE-III - Edinburgh, 25-27 October,2000.


Dutch Civil Aviation Authority, supports the identifica-tion of the relationships between causal factors. Theprotocol helps in the identification of the causal factorsfor a specific accident/incident scenario and it will alsoshow the affected elements in the ATS.

11.3.6 Collaborations

• Trinity College Dublin

The collaboration with Trinity College Dublin (TCD), inthe framework of the Leonardo Project “STAMINA”(Safety Training For Aircraft Maintenance Industry),lead to development of a specific “module” of theoverall STAMINA training course. The module was spe-cifically designed for team-related human factors issuesfor aircraft maintenance personnel.

Specifically, the delivery and exploitation of STAMINAtraining courses has now moved beyond the analysisand assessment by end-users and partners involved inthe project. A Welsh consortium called WATAG (WalesAerospace Training Advisory Group), interested in theexploitation of Stamina products, has made STAMINAwork outside the boundaries of the project itself. As such,a joint effort between TCD and ISIS is still maintainedthroughout STAMINA training delivery and exploitationacross other European maintenance organisations.

• Politecnico di Milano

Since 1992, an extensive research programme is beingcarried out in collaboration with the Aerospace Depart-ment with the objective of developing and implementinginto a simulation frame an integrated model of a flightcrew, an aircraft, and the surrounding environment.The main purpose is to make available an exploratorytool for conducting qualitative and quantitative analy-ses of normal and critical flight conditions. The latestimprovements to this exploratory tool focused on:

the Crew Model

In the previous versions of the tool, the simulationconsidered either only one pilot or the same modelwas used for the two/three crew members of the B747.

In the latest improvement of the crew model two kindsof paradigms for the human were considered andmerged. This permitted certain aspects related to bothsingle aircraft handling and co-operation and commu-nications between (three) crew members to be consid-ered. The results show that without the use of a dynamicapproach, i.e., simulation modelling, a static analysisof these procedures would not have permitted the de-tection of weak points [11.6-11.9] (Figures 11.3 and 11.4).

the Airborne Collision Avoidance Systems - ACAS

The work focused on alarms emitted by ACAS and analys-ed conflict situations between two aircraft equippedwith this system. A multi-aircraft simulator - MAS - allowsthe dynamic reproduction of the trajectories of two air-crafts: the first one simulated during the approach tolanding phase and the second one invading either thecaution or the warning area of the former. The potentialrisk of collision has been analysed through an exhaustivecollection of case studies.

The conclusions drawn are that anti-collision systemsare not always the ideal solution. Indeed, less rarely thanexpected, they turn out to be far from perfect devices andcan generate additional risk rather that improve safety.

11.3 Overall model and results of the failure to the landing gear. 11.4 Trajectory of two aircrafts and escape manoeuvre of

aircraft with TCAS.

References

11.6 Trasi, A. – Crew simulation for the analysis of critical situation in flight safety – M.Sc. Graduation Thesis inAeronautical Engineering (in Italian), Politecnico di Milano, Italy, 2000.

11.7 Barbieri, M. – A Trajectory Simulator to Study Anti-collision Alarms Emitted by TCAS Equipment, – M.Sc.Graduation Thesis in Aeronautical Engineering(in Italian), Politecnico di Milano, Italy, 2000.

11.8 Pedrali, M., Trasi, A. – Simulation and Modelling ofFlight Crew: Past, Present & Future at JRC – 19th

European Annual Conference on Human DecisionMaking and Manual Control - EAM 2000. Ispra, Italy,26-28 June 2000. pp. 81-92

11.9 Pedrali, M., Trasi, A. – Crew Cognitive Simulation forFlight Procedure Analysis – Proceedings of theInternational Conference on Human-ComputerInteraction in Aeronautics - HCI Aero 2000. Toulouse,France, 27-29 September 2000. pp. 25-30.

78

OVERVIEWM. Géradin, V. RendaThe European Commission is strongly engaged inresearch and development for the protection of theEuropean population against the consequences ofnatural and man-made hazards. A significant effort iscarried out in the field of safety in structural mechanicswith reference to buildings and bridges, means of trans-port (chapter 13) and protection of cultural heritage.The activities are performed at ISIS and take advan-tage of the unique facility in Europe – the EuropeanLaboratory for Structural Assessment (ELSA). Full/largescale tests are performed at ELSA to assess the behav-iour of structures of civil and architectural intereststricken by strong earthquakes. The work related tothe protection of cultural heritage includes the devel-opment of technologies for structural diagnostics andtechniques for strengthening/repair of the structuresvalidated through full/large scale seismic tests per-formed at ELSA. All these activities are complementedby sophisticated numerical modelling and analyses.

12.1 SEISMIC PROTECTION FOR CIVIL ANDCULTURAL HERITAGEA. Pinto

Earthquakes are by far the most deadly natural disas-ters in the world. In the last 15 years, more than 5,000people have died in earthquakes occurring in the Euro-pean Union. In 1980, a terrible earthquake struck south-ern Italy, killing 4,580 people and leaving 250,000homeless. More recently, in 1997, an earthquake inAssisi and in Athens (1999) caused extensive damageand human suffering. The death toll in Turkey (1999)was estimated to be 40,000 and more than 200,000housing units will have to be built. There is however animportant reduction in the relative number of casual-ties (causalities/inhabitants), but the cost of damageis escalating with very negative impacts (includingfinancial) on society. In fact, in spite of the importantefforts and investments made in earthquake protec-tion, the damage seems greater today for two reasons:countries are more densely populated, including thoseat risk; and, there are new industrial infrastructuresthat may be vulnerable in the event of an earthquake:gas and oil pipelines, chemical factories, communica-tion networks, etc.

How should we approach the problem? We must useall our creative skills and energy and act in four mainareas: 1) to develop and qualify appropriate designand re-design methods, incorporate new technologiesand materials with improved performance in construc-tion and retrofitting of existing vulnerable buildingsand infrastructures; 2) to provide a complete set ofregulatory documents including the most recent researchand technical achievements as well as a more rational

and accurate quantification of the earthquake actions;3) to progress towards more effective preparednessand earthquake mitigation programmes including alsocivil protection and educational aspects; 4) to involve,assist and co-ordinate the efforts of all concernedregional, national and European authorities and insti-tutions for a more rational definition and developmentof earthquake risk mitigation programmes as well asto promote international collaboration.

ELSA is acting along these lines. Specifically, there isthe on-going research project on earthquake assess-ment and retrofitting of structures, which covers threemain topics: 1) ordinary constructions (buildings andbridges), 2) masonry and built cultural heritage and3) active, semi-active and passive systems and tech-niques for structures subjected to dynamic loading(earthquake, wind, etc.). Substantial progress wasmade on all three topics with special emphasis on ma-sonry structures for which a specific sub-project wasset-up and developed during 2000 as described here-after. Topics 1 and 2 were also undertaken joining theefforts and synergies between the institutional project(entitled “Research in support to the implementationand validation of the EUROCODES; Research for struc-tural vulnerability assessment, strengthening/repairtechniques for civil and cultural heritage structure un-der earthquakes”) and several competitive projects.Examples are the work on assessment and retrofitting ofexisting constructions (buildings – ICONS and SAFERRresearch networks, bridges – VAB project) and on activecontrol (ACE and CASCO projects) are worth mentioning.

Furthermore, ELSA organized in November 2000 an in-ternational workshop on “Mitigation of Seismic Risk”to support recently affected European countries. Theworkshop aimed at a better co-ordination of the effortsfor the mitigation of seismic risks in Europe. Supportcame from the Environment DG – Civil Protection; theResearch DG; the Enterprise DG – Construction; patron-age from the European Parliament; CommissionerBusquin; Commissioner Wallström; and several Euro-pean governments. Conclusions and recommenda-tions from this workshop will certainly assist nationalauthorities and the Commission, in particular, to definetheir mitigation programmes. It is believed that theSixth Framework Programme should reflect the viewsand strategic actions as discussed in the workshop,which are fully in line with Commission policies andspecifically with the recently adopted European Re-

search Area project.

Safety of Building Structures andProtection of Cultural Heritage12


12.2 ACTIVITIES CONCERNING FIBRE REINFORCED COMPOSITES

ELSA has extended its on-going research programmeconcerning the use of fibre reinforced composites(FRCs) in civil engineering by studying two differentapplications. One concerns the generic rehabilitationof degraded reinforced concrete columns and the otherlooks at how the seismic protection of damaged rein-forced concrete buildings may be increased by wrapp-ing carbon fabric around critical elements.

12.2.1 Upgrading and Repair of

Reinforced Concrete Elements

E. Gutiérrez, W. Johansen, D. TirelliA growing problem in the housing sector in Europe isthe poor manner in which many reinforced concretebuildings built in the housing booms after the SecondWorld War have aged. At the time of their constructionit was thought that concrete was an inert material – notprone to corrosion. Unfortunately this has not provedso and, at present, throughout all the Member States,many EU citizens have directly (or indirectly as tenantsof local housing authorities) inherited a housing stock,of which a significant proportion is in need of majorstructural repair. This scenario is not as dramatic asthe structural destruction of housing caused by earth-quakes or other natural disasters, but it does have apervasive cost both socially and economically.

In the context of our institutional research, and in col-laboration with construction companies and compos-ite manufacturers involved in the repair of degradedbuildings, and in order to gain first hand experience ofthe types of rehabilitation required and the standardrepair methods used at present, ISIS has conductedfield trips to construction sites where extensive reha-bilitation work was being carried out on low-incomehousing areas. In view of the structural degradationpathologies that resulted in the reduction of the load-carrying capacity of the main columns, ISIS has conduct-ed an experimental campaign with the aim of studyingrepair techniques using FRC pre-fabricated confiningshells. The scope of the research was to determine ifthis repair technique (which is known to work well forcircular section columns) could be applied successfullyto square and rectangular cross-sections.

Three types of reinforcing material (carbon, aramid andglass fibre) were selected and tested on rectangularprisms of plain concrete. Also, given that the angle ofthe reinforcing fibre is a fundamental parameter in theconfinement capacity of the FRC shell, various types offibre lay-up (cross-ply, unidirectional and woven fabric)were selected and compared. Another parameter studiedwas the effect of the geometry of the corner radius onthe confining capacity of the shell.

Examples of the test set-up, showing two columncores confined with shells made from aramid (Figure

12.1) and carbon (Figure 12.2) fabrics, give an impres-sion of how the experiments were conducted within a360 ton hydraulic press. The experiments consisted ofmonitoring the forces and the volumetric expansion/contraction of the column sections as the load wasapplied at the ends of the concrete prisms.

Our findings suggest that FRC pre-formed shells canbe used effectively to confine concrete cores and in-crease the load carrying capacity by a factor of three.However, to obtain the best results the cavity betweenthe concrete core and the confining shells should bepressure grouted with a cement filler. Moreover, in orderto increase the confining resistance of the shells, andin analogy to pressure vessel design, the corner radiusis increased (Figure 12.1) in order to reduce the sec-ondary bending stresses at the corner (Figure 12.2)

that would otherwise generate premature delamina-tion failure.

12.1 Aramid confining shell.

12.2 Square section

carbon fibre shell.

80

12.2.2Application of Composites to Seismic

Resistance of Reinforced Concrete Buildings

P. Negro, A. ColomboThe improved confining capacity of composite materialsdescribed above for the generic case of vertical load-ing has been exploited to enhance the seismic responseof non-ductile reinforced concrete structures. A reinfor-ced concrete structure damaged by previous experimen-tal activities has been used to assess the effectivenessof the wet lay-up application of carbon fibre-reinforcedpolymers (CFRP) as a repairing technique (Figure 12.3).The investigation, carried out in close collaborationwith industry, aimed at verifying the efficiency and thefeasibility of this intervention on a structure as a whole.

The anisotropic properties of CFRP have been exploitedin repairing the different type/s of damage characteris-ing the critical zones of the building. After the injectionof all cracks with epoxy resin, mono-directional carbonfibre wraps were glued around all the member (beamsand columns) ends. The shear walls were wrappedthroughout the height of the building using quasi-iso-tropic (fibres oriented at 0, 90 and ± 45 degrees) wraps.After the intervention, the frame was subjected to thesame input motion applied during the previous experi-mental campaign. A preliminary analysis of the resultsseems to demonstrate that, because of the interven-tion, the loss in strength caused by the damage suf-fered during the previous tests was recovered. In addi-tion, the capacity of the frame to behave in a ductilefashion appears to have improved.

12.3 SEISMIC ASSESSMENT OF MASONRYSTRUCTURESP. Pegon, Y. Le Pape

This activity is intended (1) to allow the calibration/development of advanced nu-merical models for ma-sonry with particular emphasis on the in-plane cyclicbehaviour and (2) to provide accurate guidelines forretrofitting interventions after a seismic event by as-sessing the efficiency of traditional and innovative re-pair strengthening techniques. Connected numericaland experimental works have been carried out. Theobjective of the numerical work is to provide a com-plete study of the in-plane behaviour of masonry wallsfor various mechanical scales. It is heavily connectedwith the experimental work. Three scales are consid-ered: microscopic, macroscopic and structural withparticular emphasis on the existing connections be-tween the corresponding modelling strategies. Indeed,the modelling of masonry generally falls between twodistinct classes:• the micro-mechanical models (sometimes referred

to as discrete models) accounting for the morphologyof masonry:

• the macro-mechanical models (also called homoge-neous or continuous models) account for the behav-iour of a typical relevant unit cell by establishing adirect constitutive law between the average stressesand strains states.

The micro-mechanical approach is seen as the bestmodelling scale to understand the salient features ofthe in-plane behaviour of masonry panels. The intro-duction of a tensile cut-off within the joint model hasbeen proved to be a key point in the modelling of ma-sonry. The ratio of tensile strength with respect to thecohe-sion is a factor of great influence upon the behav-iour of a shear wall. Nevertheless, such modelling cannotreasonably be applied to deal with a structure largerthan a single wall. For instance, the modelling of com-plete infilled-storey buildings definitely requires adopt-ing another standpoint at a macro-scale level.

The macro-mechanical models shall be obtained byeither adopting a phenomenological standpoint assum-ing the masonry as a “no-tension material”, or usinghomogenisation techniques. Obtaining a consistent con-tinuous anisotropic inelastic model has been provedto be a difficult task due to the materials intrinsic prop-erties and the lack of comprehensive experimentalresults. Therefore, one of the objectives of the institu-tional project is to complete an accurate database ob-tained both by numerical and experimental means, soas to provide information about the mechanical behav-iour at different scales

12.3 Wet lay-up application of a quasi-isotropic carbon FRP

on the shear walls of a reinforced concrete dual frame.

Three types of numerical tests are considered (Figure12.4): (a) at the structural level, in-plane cyclic behav-iour of a shear wall, (b) at a “mesoscale”, large as-semblages, also called “wallettes”, with inclined bed-joints are tested under biaxial loading conditions, and

finally, (c) at the lowest scale, the behaviour of a basiccell under the assumptions of periodic medium is alsostudied with a view to deriving the mechanical globalconstitutive law through the homogenisation tech-niques. It should be noted that (c) can only be testedby numerical means due to the impossibility of mak-ing an adequate testing rig cabable of reproducing theperi-odic conditions. Tests (b) - Figure 12.5 - and (a) -Figure 12.6 - are also performed by experimentalmeans thanks to ELSA’s full-scale testing facilities.The multi-scale ap-proach proposed in the numericalframework requires collecting experimental data forevery studied scale. Up to now a single test of eachkind has been performed.

Some future experimental tests will be performed soas to provide a significant feedback to the numericalmodelling of unreinforced masonry.


Shear wall LargeAssemblage

ElementaryCell

(a) Structural scale

(b) “Mesoscale”

(c) “Microscale”

12.4 Scales of modelling

12.5 Large assemblage/“walette” test set-up - structural behaviour.

12.6 Shear wall test set-up - structural behaviour.

Horizontal displacement [mm]

Hor

izon

tal l

oad

[kN

]

60

40

20

0

-20

-40

-60-20 -15 -5-10 0 5 10

0

-50

-100

-150

-200

-250

-300

-350

-400

-450

Com

pres

sion

load

[kN

]

Horizontal displacement [mm]

-5 -4 -2-3 -1 0

82

12.4 FURTHER DEVELOPMENTS IN DYNAMICCONTROL OF EARTHQUAKE ENGINEERINGFACILITIES (FUDIDCOEEF)G. Magonette

The Continuous Pseudo-Dynamic (PsD) testing methodused at ELSA is implemented by means of a synchro-nous process with short control period (1 ms) andsmall integration time step. This introduces somechallenging difficulties for the implementation of thesub-structuring technique because if the analyticalstructure is complex, the analytical process is unableto perform even an elastic computation during a con-trol period of the experimental process. Different timesteps must be used by the two processes. To face thisproblem an improved inter-field integration schemehas been developed. This scheme, denoted as “inter-field with correction in force and in displacement”, usesdifferent time steps for the analytical and experimentalprocesses and provides a better solution by doublingthe integration process to solve the analytical part[12.1-12.2].

Several experimental tests have been carried out on asimple structure to validate an implementation includ-ing the explicit CD scheme in the experimental processand the implicit trapezoidal integration rule in the ana-lytical process. Further tests are in preparation on alarge steel structure (Figure 12.7).

12.5 ACTIVE AND SEMI-ACTIVE CONTROL IN CIVIL ENGINEERING (ACE AND CASCO)G. Magonette

An active control system is one in which an externalsource powers control actuator(s) that apply forces toa structure in a prescribed manner. These forces canbe used to either add or dissipate energy in the struc-ture. The methodology considered in the ACE project isbased on active tendons in which the damping is in-duced by the control of the displacement of the cableanchor point. This concept developed at the UniversitéLibre de Bruxelles has a strong physical support andits effectiveness has been fully confirmed experimen-tally by numerous tests performed on a large-scalemodel of a cable-stayed bridge [12.3-12.5]. As an alter-native to actively controlled actuators, semi-active (S-A)devices offer some attractive properties: unconditionalstability, low power requirements (which is of greatimportance in case of an earthquake) and efficiencycomparable to active devices. The two concepts arecompared in Figures 12.8 and 12.9. The particularity ofS-A control comes from the fact that it is not possibleto directly command the S-A damper to generated aspecified force f, because the response of the dampernot only depends on the input command (valve strokefor a variable-orifice device, applied magnetic field fora magneto-rheological damper), but also on the in-stantaneous motion of the piston. However, the forceproduced by the S-A damper may be increased or de-creased by adjusting the value of the input command.An important limitation comes from the fact that onlydissipative control forces can be generated. So, mostoften, the S-A strategy consists of a modified activecontrol law, illustrated in Figure 12.9, where the S-Ainput command is determined by “clipping” the activecontrol strategy.

The work currently carried out in the CASCO projectaims at adapting the active system presented aboveinto a S-A system. Therefore, some important techno-logical issues have to be addressed: (i) dealing with astatic load in the S-A damper, (ii) limitation of the in-ternal friction, (iii) minimum required stroke for the S-Adevice to be efficient.

12.7 Large steel frame used for the validation of PsD method

implemented with the inter-field integration scheme.

References

12.1 Magonette G., Pegon P., Buchet P., Bonelli A.– Continuous Pseudodynamic Testing with Non-Linear Substructuring – Second EuropeanConference on Structural Control (2ECSC). ENPC,Champs-sur-Marne, France, 3-6 July, 2000.

12.2 Magonette G., Pegon P., Pinto A.V. – The New ELSALaboratory – Proceeding of Belgirate Workshop on“Mitigation of Seismic Risk Support to RecentlyAffected European Countries”, 27-28 November,2000 Belgirate, Italy.

12.6 VULNERABILITY ASSESSMENT OF BRIDGES (VAB)A. Pinto

ISIS is participating in the VAB (Advanced Methods forAssessing the Seismic Vulnerability of Existing Motor-way Bridges) project as the major partner for the test-ing activity. Four large-scale models of typical bridgepiers with different span ratios were constructed [12.6]and will be tested at ELSA using non-linear substruc-turing techniques, which allow the online numericalsimulation of the behaviour of part of the bridge piers,combining this numerical simulation with the physicaltesting of the most critical piers.

Insufficient consideration for seismic risk in bridge de-sign has resulted in inadequate detailing of confiningsteel and insufficient shear reinforcement in the bridgepiers, insufficient seat length of bearings, and inade-quate design of the bridge abutments. There are manyopen questions concerning the ductile behaviour oflarge bridge piers, in particular those with a rectangu-lar hollow cross-section commonly used in Europe.

The VAB project aims at contributing to the solution ofmost of these aspects [12.7], in particular, by identify-ing methods to assess vulnerability of existing bridgesand by developing appropriate retrofitting solutionsand techniques for bridge piers with a rectangular-hollow cross-section.

Further development of the non-linear substructuringto be used in the online control of the experiments wasachieved in 2000, combining this test technique withthe continuous pseudo-dynamic testing technique[12.8], which allows for more accurate experiments. Finaldefinition and design of the testing and measuringsystems and devices were carried out and the instru-mentation plans were proposed to the project partners.A discussion was launched on the seismic retrofittingsolutions and techniques for bridge piers with a rec-tangular hollow cross-section, which will assist theretrofit of one squat pier to be tested at ELSA.

fcSemi-activedamper adjustment

(b) Sensor

M

Semi-activedamping law

Activedamping law

f


Pump

Active hydraulic actuator+ local controller

(a) Sensor

M

Activedampinglaw

~f fc~

12.8 Active control. 12.9 Semi-active control.

References

12.3 Magonette G., Marazzi F., Preumont A., Bossens F.,Auperin M., Dumoulin C. – Active Control of Cable-Stayed Bridge: Concept and Large-Scale Experi-ments – Proceeding of MOVIC-2000, University ofTechnology, Sydney, 4-8 December 2000.

12.4 Magonette G., Marazzi F., Preumont A., Bossens F.,Auperin M., Dumoulin C. – Active Control of CivilStructures: Theoretical and Experimental Study –Proceedings of IMAC-XIX: Conference on StructuralDynamics, Orlando, 5-9 February 2001.

12.5 Dumoulin C., Försterling H., Magonette G., Bossens F. – Active control of cable-stayed bridges: large-scaleexperiments – Proceedings of 1st IFAC-Conferenceon MECHATRONIC SYSTEMS, Darmstadt, 18-20September 2000.

References

12.6 Geradin, M., Pinto A.V. (editors) – Mitigation of Seis-mic Risk – Support to Recently Affected EuropeanCountries. Workshop Proceedings – 27/28 Novem-ber 2000, Belgirate (VB), Italy. Special PublicationNo. I.00.114 – EC, JRC-ISIS, Ispra, Italy (2000).

12.7 Flesch, R. et al. – Advanced Methods for Assessingthe Seismic Vulnerability of Existing MotorwayBridges. In “Mitigation of Seismic Risk – Support toRecently Affected European Countries”, Geradin andPinto (editors). Special Publication No. I.00.114 – EC,JRC-ISIS, Ispra, Italy (2000).

12.8 Pegon, P., Magonette G. – Continuous PSD Testingwith Non-linear Substructuring: Recent Develop-ments for the VAB Project. Special Publication No.I.99.142 – EC, JRC Ispra (1999).

12.7 ANCHORAGES IN CONCRETE UNDERDYNAMIC LOADINGG. Solomos

Anchorage systems in the concrete construction in-dustry are given special attention, which is justified bythe following global trends: increased interest inearthquake retrofitting world-wide; increased interestin preservation of the historical built environment andin the re-use and rehabilitation of existing structures,particularly in Europe; increasing use of pre-fabricatedstructural elements; other special anchorage demands(offshore construction, road safety barriers, nuclearpower plants etc.).

Understanding of the anchorage response, principallyto dynamic loads, is the principal objective of this ac-tivity. New materials design of high performance con-crete (with steel fibres), materials modelling (includingthe strain rate dependence of the stress-strain curveof concrete and its softening branch) and numericalsimulation tools have been pursued. An extensive ex-perimental programme has supported all theoreticaldevelopments. Central to this high strain-rate testingactivity, has been the employment of various Hopkinsonbar techniques (mainly the Large Dynamic Test Facility- LDTF), particularly suited for precision pull-out tests.

Specimens have been made of normal and high per-formance concrete and include: plain cubic specimens[12.9], rebars (cast-in-place and post-installed) andanchors (of chemical, undercut and headed stud type)– Figure 12.10. A large number of tests has been per-formed at the LDTF and interesting results have beenobtained for the dynamic behaviour of 20 mm diame-ter rebars and anchors of diameters of up to 16 mm –Figure 12.11. Parallel material modelling work is carriedout for concrete along the lines of the lattice, micro-plane and strain gradient plasticity models.

The big part of this activity [12.10] is performed withinthe framework of the project ANCHR “Anchorages inNormal and High Performance Concretes subjected toMedium and High Strain Rates”. Other partners include:DENSIT a/s (Denmark), HILTI (Lichtenstein), ENEL-CRISItaly), Politecnico di Milano (Italy), University of Patras(Greece) and Bekaert (Belgium).

12.8 THE ICONS AND SAFERR RESEARCH NETWORKSA. Pinto

ISIS is participating very actively in the training pro-grammes of the European Commission. Since 1994,ELSA has been involved in a) the “Access to Large-scale Facilities” progamme by opening its doors to ex-ternal researchers, b) in research training networkswith research topics focused on earthquake engineer-ing and also c) in Marie Curie fellowships by hostinga few young researchers who have developed theirresearch projects at the facility.

Recently, ELSA was considered a European Marie Curietraining site, which enhances the key role of ELSA inthe European earthquake engineering community.Furthermore, ELSA was considered and financed as alarge-scale research infrastructure integrating the Eu-ropean Consortium for Large Earthquake and DynamicEngineering Research (ECOLEADER) [12.11] and a newgrant was awarded for the continuation of its partici-pation in the SAFERR research training network [12.12].There is a very important feature that makes these

84

12.10 Concrete blocks 64x64x28(cm) of the headed studs of

Ø16mm tested under dynamic pull-out conditions.

140

120

100

80

60

40

20

0

Forc

e (k

N)

Displacement (nm)

M16 Headed Studs

dynamic D2dynamic D2dynamic D1dynamic D1quasi-staticquasi-static

0 2 4 6 8 10 12 14 16 18 20

12.11 Characteristic force-displacement curves of M16

headed studs under quasi-static, low-dynamic and

higher-dynamic pull-out testing.

References

12.9 Ulfkjaer, J., Labibes, K., Solomos, G., Albertini, C.– Tensile failure of normal concrete and steel fibrereinforced concrete at high strain rates – Proceed-ings Fracture Mechanics of Concrete Structures,FRAMCOS-3, Gifu Japan, 12-16 October, 1998.

12.10 Sørensen, E., Solomos, G. – Anchorages in normaland high performance concretes subjected to medi-um and high strain rates – 4th Annual Workshop,Targeted Research Action, Environmentally FriendlyConstruction Technologies, Brussels, Oct. 2000.


afore-mentioned projects effective and visible to theinternational community outside Europe (e.g. USA andJapan). The synergies between the experimental facili-ties and the European earthquake engineering com-munity, including young researchers financed underthe research networks and the Marie Curie grants, ena-bled a series of topics relevant to seismic protection ofconstructions to be tackled. In particular, the key con-tribution for the calibration and further developmentof Eurocode 8 (EC8) should be underlined.

The ICONS (Innovative Seismic Design Concepts forNew and Existing Structures) project covered five re-search topics, namely: 1) Advanced methods for thequantification of the seismic action; 2) Assessment,retrofit and repair of existing buildings; 3) New defor-mation-based design (DBD) methods; 4) Compositestructures, and; 5) Shear-wall structures. Importantachievements are summarised as follow: a) A new setof displacement spectra to be included in EC8, whichallow application of DBD methods; b) Calibrated rapidscreening methods for seismic assessment and quali-fication of strengthening solutions and techniquessupporting calibration of EC8 - part 1.4 - “Repair andStrengthening”; c) Development of DBD methods andprocedures for design and re-design of structures inearthquake prone zones. A promising, truly European,method was developed under ICONS and proposed tothe international scientific community; d) Researchunder composite structures allowed the inclusion ofthe corresponding annex of EC8 as a normative part; e)Research on shear-wall structures led to several testson U-shaped walls carried out on a shaking-table (CEA-Saclay) and cyclic tests at ELSA, which constitute aunique set of data for design of non-rectangular shearwalls and for model calibration.

The SAFERR (Safety Assessment for Earthquake RiskReduction) project intends to continue to enhance theearthquake engineering research in Europe followingan approach based on three underlying themes[12.13], which are: advancement, application and dis-semination of measures for reducing seismic risk inEurope. Thus, emphasis has been placed on advancingthe state-of-the-art in assessment of hazard (exposure)through task 1 “Characterisation of Seismic Hazard”.The perpetual problem of how to design and buildsafely and economically in areas of low seismic expo-sure (which covers a large proportion of Europe) andwhere taking no precautions may lead to extensivelosses in small-to-medium earthquakes is dealt with intask 2 “Design and Assessment in Low SeismicityRegions”. Whilst appreciating the value of developingnew methods of design for new constructions, the ex-isting infrastructure stock remains highly vulnerable inmany parts of Europe. Therefore, task 3 addresses

“Strategies and Techniques for Vulnerability Reduction”,which deals with a selected number of sub-topics inassessment and retrofitting of buildings, bridges andtheir foundations. Finally, integrating the above hazard-related and vulnerability-related activities in a usable“risk” framework is undertaken in task 4 “Risk Assess-ment Systems”.

References

12.11 Severn, R. – Earthquake Engineering Research Infra-structures. In “Mitigation of Seismic Risk – Supportto Recently Affected European Countries”, Geradinand Pinto (Eds.). Special Publication No. I.00.114 –EC, JRC, Ispra, Italy (2000).

12.12 Elnashai, A. – Research and Training Networks inSupport of European Seismic Risk Mitigation. In“Mitigation of Seismic Risk – Support to RecentlyAffected European Countries”, Geradin and Pinto(editors). Special Publication No. I.00.114 – EC, JRCIspra, Italy, 2000.

12.13 Pinto A.V. – Introduction to the European researchprojects in support of Eurocode 8 – Proceedings ofthe 11th European Conference on EarthquakeEngineering. Balkema, Rotterdam, (1998),ISBN 90 5410 982 3.

12.12 Construction of the large-scale bridge piers outside ELSA.

12.9 PROJECT LADIB (LABORATORIO DINAMICOINDUSTRIALE DELLA BASILICATA)G. Magonette, P. Negro

As a part of the contractual activities for the Basilicataregion, ELSA was given the task of providing referenceexperimental results of pseudo-dynamic tests to be usedin the commissioning of the new structural mechanicslaboratory at Potenza, Italy. Two complete structureswere built and are being tested. The first is a reducedscale model of a steel building with moment-resistingwelded connection (Figure 12.13). This will allow dif-fering pseudo-dynamic tests to be conducted, both inthe elastic and non-linear ranges, to be reproduced inthe new laboratory in Potenza. The comparisons willbe extended to the real-time dynamic tests conductedat the ISMES shaking table on a similar structure as apart of the STEELQUAKE project [12.14].

The second structure is a three-storey reinforced con-crete building designed according to old seismic designpractice. Beside the data needed for the commissioningof the new laboratory, the tests on this structure willthrow light on the behaviour of flat-slab structures, forwhich many design principles enforced by the currentdesign codes are violated.

12.10 TUNNELLING PROJECTP. Negro

The objective of the Tunnelling project [12.15] was todevelop a monitoring system, based on a novel fibre-optics technology, to detect the onset and progressionof damage in the concrete lining of transportation tun-nels. During the first phase, ELSA produced the experi-mental results needed for the definition of an energy-based damage model for these structures. In 2000,tests were conducted to assess the performances of thefibre-optic system and of the damage evaluation model.The fibre-optic sensors were mounted on the surfaceof the specimen, and interfaced by multiplexed con-nection to the optical acquisition unit. The specimen(Figure 12.14) underwent a deformation time-historyrepresentative of the seismic action during an earth-quake, and the measurements were compared withthose made by means of traditional sensors (Figure

12.15). Finally, the damage assessments made by theenergy-based damage model were compared with theobserved damage states.

86

References

12.14 Bairaktaris D., Frondistou-Yunnas S., Kalles D.,Stathaki A., Kallidromitis V., Kotrotsios G., Negro P.,Colombo A., – Intelligent Monitoring of SeismicDamage in Reinforced Concrete Tunnel Linings –Workshop on Reducing Earthquake Risks toStructures and Monuments in the EU, Cambridge,UK, 26-27 November 1998.

12.15 Taucer F., Negro P., Colombo A. – Cyclic and PsDTesting of a Moment Resisting Steel Frame withBeam-to-Column Welded Connections – Journal ofEarthquake Engineering, Vol. 4, No. 4, 437-477,2000.

12.13 Reduced-scale model of a steel building with moment-

resisting welded connections.

12.14 Experimental set-up for the tests on a portion of

reinforced concrete tunnel.

12.15 Fibre-optic sensors mounted in parallel with traditional

sensors.


12.11 3D RECONSTRUCTIONV. Sequeira, E. Wolfart, J.G.M. Gonçalves,P. Dias, M. Ruggieri

In the last years, ISIS has increased its expertise in 3Dreconstruction by collaborating in competitive projects.The objective of 3D Reconstruction is to build a 3D photo-realistic computer model of an environment “as-built”,i.e., with all its visual and spatial characteristics. Thistechnique applied to culture heritage results in visuallyappealing 3D models to make the cultural site accessi-ble over the Internet (Virtual Heritage), as well as foraccurate 3D measurements for documentation andconservation purposes (Cultural Conservation).

The RESOLV project aimed at the development of thebasic 3D reconstruction technologies, whereas INFOBOYwas more application oriented, specifically for adver-tising old farmhouses in rural tourism. Figure 12.16 dis-plays the 3D model of a heritage farmhouse. The mod-els are inserted inside a virtual sphere “painted” withthe surrounding photographs. This technique consti-tutes an effective way to provide spatial context tomodels of buildings and monuments.

At the end of 2000, BBC (British Broadcasting Corpora-tion) invited the JRC to model the “Sala dello Scrutinio”(45m x 15m x 10m) at the Doges Palace, in Venice, Italy.It took four hours to scan the room, which is normallyclosed to the public and contains major paintings suchas “Giudizio Universale” by Palma il Giovane. Figure

12.17 shows some snapshots of the model illustratingdetails of the room as well as BBC’s Peter Snow in twoversions: photograph and 3D avatar.

12.16 Three-dimensional reconstructed model of a farmhouse

including the background for spatial context.

12.17 Snapshots of the 3D model of the “Sala dello Scrutinio”,

Doges Palace, Venice, Italy. Two representations of

BBC’s Peter Snow: photograph and 3D avatar inside

the model.

88

OVERVIEWM. GéradinThe experimental work achieved in the field of safetyin transient structural dynamics consists of impact andcrash tests on material samples and structural compo-nents. The large-scale tests are performed at the LargeDynamic Test Facility (LDTF), a unique facility in Europe.The smaller scale tests are performed in the DynamicMaterial Testing Laboratory. All these tests are com-plemented by sophisticated numerical modelling andanalyses, which are indispensable to assess the struc-tural integrity of a large variety of systems and instal-lations. In particular, during 2000:

• An important effort has been made in the numericalsimulation of fast transient dynamic phenomenaoccurring in complex fluid-structure systems sub-mitted to impulsive loading. For this purpose, thecollaboration with CEA (Commissariat à l’EnergieAtomique) has been intensified to integrate the lat-est expertise of both research teams in the commonsoftware platform EUROPLEXUS. This collaborationis also being extended to other European partners.

• The experimental activity dealing with certain pas-sive safety aspects in the field of transport, wherecrash and impact are involved, has been particularlyintense. This work has been planned and is intendedto provide support to the Enterprise DG in terms ofthe Commission’s normative action policies. The as-pects concerned are: frontal impact of vehicles andassociated crashworthiness, road safety barriers andpedestrian friendliness of car fronts.

• An in-depth study of constitutive models of variousmaterials subjected to impact loading was also start-ed with the objective of bringing the constitutivemodelling activities up to the state-of-the-art levelin modern material modelling. The numerical simu-lation of notched specimens subjected to transientloading has been started.

13.1 NUMERICAL MODELLING - EUROPLEXUSF. Casadei, T. Dyngeland, J.P. Halleux

ISIS’s activities regarding the numerical simulation offast transient dynamic phenomena occurring in com-plex fluid-structure systems submitted to impulsiveloading have now evolved into a large scale actionaiming at bringing together similar research work per-formed elsewhere. In this respect, CEA and the JRChave already launched a five-year collaboration agree-ment effort for the setting-up of a new computer code,EUROPLEXUS, which effectively replaces the formercodes CASTEM-PLEXUS and PLEXIS-3C. EUROPLEXUSis expected to offer features, both in terms of actualcomputational mechanics models and software tools,attractive enough to convince others to join in this com-mon development effort. In order to ensure an effec-

tive usage of such research efforts, in particular at theindustrial level, commercialisation through an interna-tionally recognised finite element software vendor iscurrently being negotiated.

A collaboration agreement between ISIS and EDF(Electricité de France) was established in October,which will run for 16 months to the end of January2002. The topic of the collaboration is “Fast Dynamicsand Impacted RC-Structures” and includes one secondedperson from EDF working at ISIS for the full period of16 months. The collaboration project is well under waywith the first main topic related to the development ofa global reinforced concrete model for finite shell ele-ments. At the end of December, a document describingthe integration principles for two different integrationmethods was finished. Furthermore, the two integra-tion methods have been implemented in EUROPLEXUSand are now being tested. The material model involvedin the work is based upon a global moment-curvaturelaw and with uncoupled membrane and bending effectsat this level of development.

13.1.1 Industrialisation of FEM Advanced Software

for Large-scale Fluid-structure

Transient Dynamics Applications

The EUROPLEXUS documentation has been arrangedin a CD-ROM for easy presentation and access to theuser manual and various articles, reports, slides, etc.that are available for the EUROPLEXUS software. ISIS’scontractor, Socotec Industrie, has forwarded a finalversion of the CD-ROM and the product is currentlyunder evaluation at ISIS.

A prototype version of the EUROPLEXUS preferencesoftware (user interface based on Samcef/Design) hasbeen developed by Socotec Industrie in collaborationwith Samtech and is now being evaluated. The part ofthe EUROPLEXUS preference software that shoulddeal with the post-treatment of the numerical resultsis not developed yet. A meeting next January/February2001 will set the path for finalising the last part of thissoftware. The prototype version is developed for theWindows NT operating system for PCs.

13.1.2Report on Phase 1 of the EUROPLEXUS

Agreement with CEA

The EUROPLEXUS project (between CEA and JRC,signed in December 1999) consists of two phases: (1)creation of the initial version of the code by suitablymerging its two ancestors CASTEM-PLEXUS (CEA) andPLEXIS-3C and (2) fur-ther common development. Phase1 was started at the beginning of 2000 and may be con-sidered complete. The major achievements obtainedduring this phase are briefly described hereafter.

Safety in Transient Structural Dynamics13


• Common software development environment

The merging activity has been performed by meansof a new common software development environ-ment that allows synchronised and harmonious evo-lution of the code (including its source, benchmarktests and basic documentation) by several develop-ment teams, located at ge-ographically distinct sites,and using different computer platforms. The envi-ronment is based upon communication through theWorld Wide Web, and guarantees that the same,common and unique version of the code (versionunder development) can be used at all devel-opment sites (“mirror sites”). The collection, testingand distribution of the evolution packages is per-formed by a “central site”, which is accessiblethrough the web. This environment is entirely basedupon public domain software utilities and is readilyextendible according to user needs [13.1-13.3]. Itsdevelopment has been active since January 2000and has proven to be useful and effective for thepurpose.

• Code merging

The two EUROPLEXUS ancestors, CASTEM-PLEXUSand PLEXIS-3C, were “frozen” in their status of Dec-ember 1999. Then, the first version of EUROPLEXUSwas prepared by gradually merging all features andmodels from PLEXIS-3C into CASTEM-PLEXUS.

• User’s manual

The EUROPLEXUS manual is part of the commonsoftware development environment and is constantlydeveloped together with the code. The English lan-guage and the LaTeX format have been adopted.Various formats are automatically produced by theevolution procedure: DVI and PDF formats (suitablefor exchange and on-screen visualization), Post-Script format (suitable for printing), and HTML for-mat (for on-line browsing).

• Source code

The merging of the source code has been completedso that the current version (as of end December2000) may be considered to be the initial version ofEUROPLEXUS. It contains virtually all the modelsand features of both its ancestors. During the merg-ing activity, many obsolescent and old-fashionedcode parts have been cleaned up. The programmesource currently consists of 325,000 code lines col-lected into more than 2,000 source files.

Figures 13.1, 13.2 and 13.3 present some novel applica-tions that are now possible thanks to the EUROPLEXUScollaboration with CEA, namely crash analysis, high-speed impact with fragmentation and fluid structureanalysis of pipelines.

13.1 EUROPLEXUS simulation of a cylindrical tube crash.

Calculation (left) experimental (right) [Courtesy of CEA].

13.2 EUROPLEXUS simulation of a bird impact on a plane

engine by the SPH method [Courtesy of CEA and Snecma].

13.3 EUROPLEXUS simulation of a transient fluid-structure

interaction in a pipeline [Courtesy of CEA].

90

13.1.3 Benchmark Tests

Another essential component of the EUROPLEXUSsystem is the benchmark tests suite. These tests areaimed at ensuring that code modifications do not con-flict with previous models (“non-regression tests”)and are an essential component of Quality Assurance.While CASTEM-PLEXUS was already endowed with aseries of such tests, these did not exist in the PLEXIS-3Cenvironment, therefore many new tests have been setup in order to cover as broad a spectrum as possible ofthe new features inherited from PLEXIS-3C. Over 150tests have now been implemented and have been exe-cuted at each code evolution.

13.1.4Documentation

The re-organisation of the available technical docu-mentation concerning EUROPLEXUS and its ancestorshas also been started. To date, over 80 documents(technical notes, reports, papers, etc.) are available inPDF format and may be consulted on-line via the EU-ROPLEXUS development environment.

13.1.5 Data Structure Modernisation

An important aspect of the EUROPLEXUS project is themodernisation of code programming aspects. Ances-tors were essentially written in Fortran 77. While newdevelopments are of course expected to be directly im-plemented in Fortran 90, it is also foreseen to gradual-ly convert the whole code to the new standard. Thiswill allow substantial improvements in, for example,the elimination of dimensioning directives, the optimi-sation of data structures and algorithms, etc.

Some preliminary considerations and guidelines havebeen defined [13.4]. As a practical exercise, conversionof the materials data structure has been already start-ed [13.5]. This activity is a long-term one and will mostprobably require several years to complete.

13.2 STRUCTURAL ASPECTS IN TRANSPORT SAFETY G. Solomos

13.2.1 Crash Safety Enhancement of

Vehicles by Precision Impact Tests

A significant obstacle to the use of numerical crashsimulations as an alternative/complementary test tothe present European Standard certification car crashtests [13.6] is the lack of knowledge concerning theaccuracy of the numerical predictions. For a reliableassessment of this accuracy, it is necessary to measurethe stress waves propagating inside the car structureduring a crash, which constitute the governing phe-nomenon in the case of impact loading. The LargeDynamic Test Facility (LDTF) of the JRC/ISIS has thepotential and lends naturally itself to this type ofmeasurements. Thus, a study has been initiated aimingat investigating the local behaviour of motor vehiclestructures in the event of a frontal impact, in order toprovide an independent validation of numerical pre-dictions by precision experimental techniques basedexactly on stress wave propagation considerations.The major part of this work has been performed as asupport to the Enterprise DG standardisation policies,and in collaboration with REGIENOV (RENAULT Re-cherche et Innovation). The first phase included theinvestigation of the local behaviour of the front of theautomotive “body-in-white”, in the event of a frontalimpact without offset. According to REGIENOV’s pref-erence the automotive body of its current Renault Cliomodel was chosen to be subjected to the precisionimpact tests of this study.

Central to this activity has been the development andemployment of the “energy flow transducers”, whoseunderlying principle has been extensively tested onsingle car components, like longitudinal beams [13.7].In Figure 13.4, these transducers are representedschematically by the elements 1, 2, …7. They are alu-minium bars and they have been selected according tostructural details concerning their cross-sectional area. Matching of acoustical impedances is a crucial factorin this case [13.8]. The cutting section of the “body-in-white” has been selected slightly distant from thefoot-well. This position presents certain advantagesfor the experiments, as the structural components ofthe car take a more linear form in that region, and thusfavour a predominantly uniaxial wave propagationpattern. These components are: two A-pillars with asmall roof section, two side-members under the floor,two sills, and the tunnel. A view of the actual experi-mental set-up is shown in Figure 13.5, where furtherdetails of the structural specimen and of the input andoutput (energy transducer) bars are seen [13.9]. All ex-periments have been performed with an impactorspeed of approximately 15m/s. In order to obtain this

References

13.1 Bung H., Casadei F., Halleux J.P., Lepareux M.– Specifications for the Multi-Site Development ofthe EUROPLEXUS Computer Code – Technical Note(forthcoming).

13.2 Bung H., Casadei F., Lepareux M., Halleux J.P. – Atelier Logiciel de EUROPLEXUS – CEA Report(forthcoming).

13.3 Casadei F., Halleux J.P., Bung H., Lepareux M.– Organisation of the EURO-PLEXUS Mirror Site(Windows NT) at JRC Ispra – Technical Note(forthcoming).

13.4 Casadei F., Halleux J.P., Bung H., Lepareux M.– Some Tentative Guidelines for the Development ofthe EUROPLEXUS Software System – Technical Note(forthcoming).

13.5 Casadei F. – A Proposal for the Reorganisation of theMaterials Data Structure in the EUROPLEXUS Soft-ware System – Technical Note (forthcoming).


velocity, a pretensioning of the LDTF cables of 260tons was applied. In addition to the strain gauge meas-urements, a fast speed camera was employed at a rateof 4,000 frames per second. A total of three tests wereperformed under the same loading conditions andwith satisfactory repeatability (Figure 13.6).

The curve of Figure 13.7 shows the force at the end ofthe incident bar at the joint with the impactor versusthe impactor displacement. This is in principle theforce-shortening diagram of the car “body-in-white”front. There is a first peak of approximately 140 tonsfollowed by peaks/troughs, which correspond to suc-cessive collapse mechanisms of the front structure.The Finite Element (FEM) code RADIOSS has beenused by RENAULT for the numerical simulations. Theexisting FEM model of the Clio car has been adapted tothe needs of the ISIS experiment by introducing appro-priate initial and boundary conditions.

• Comparison of the experimental and simulation re-sults is conducted on the basis of the force-timecurves determined for each of the structural mem-bers examined. An overall satisfactory agreementhas been observed. Both the trends of the corre-sponding curves and their numerical values are veryclose (Figures 13.8 and 13.9).

• This is a significant and encouraging outcome andjustifies the approach followed. It shows that nu-merical codes, properly validated, can deliver relia-ble results in the crash design problem. It furtherdemonstrates that numerical simulations and preci-sion experiments can help each other in complexstructural problems.

References

13.6 Directive 96/79/EC of the European Council of 15Dec 1996, on the protection of occupants of motorvehicles in the event of a frontal impact and amend-ing Directive 70/156/EC; OJ N. L 18/7, 21.1.97.

13.7 Hanefi, E. H., Wierzbicki T., Albertini, C. – Calibrationof Impact Rigs for Dynamic Crash Testing – ReportEUR 16347 EN, 1995.

13.8 Albertini, C., Labibes, K., Solomos, G. – PrecisionMeasurements of Crash Energy Absorption: fromMaterial Specimens to whole Vehicle Body – EUR17293EN, 1997.

13.9 Solomos, G., et al. – Study of local behaviour ofmotor vehicle structures in the event of frontalimpact/validation of numerical predictions –Final Report for DG-ENTR, Dec.2000.

13.4 Sketch of the car crash testing at the LDTF.

1-2-3-4-5-6-7 TRANSMITTED HOPKINSON BAR BUNDLE TO MEASURE SPATIAL CRASH RESISTANCE AND DEFORMATION OF THE CAR BODY

INCIDENT HOPKINSONBAR TO APPLY ANDTO MEASURE LOADOF THE CAR BODY

STRAIN GAUGES7

5

64

3

2

1

4

13.5 Details of output bars and marking of the vehicle

and input bar.

13.6 Side view after test 1 presenting the typical

deformation pattern of the structure after crashing.

13.2.2 Road Barrier Deformation Tests

Safety barriers are designed to restrain and redirecterrant vehicles, according to specified performancelevels of containment, for the benefit of both the occu-pants and of other road users. The requirements thatthese road barriers have to meet for certification pur-poses are set in the relevant standards, e.g. EN 1371-1-2,prepared recently by the Technical Committee CEN/TC226. Full-scale tests involving barriers as well asvehicles are prescribed in them. A significant impedi-ment to developing new road barrier systems is thehigh cost of performing these full-scale tests. Whilesome full-scale crash tests will always be needed todemonstrate that the devices work correctly, it is highlydesirable that numerical computer simulations enableengineers to reliably evaluate alternative designs morequickly and at less cost. This can help to drastically re-duce the number of crash tests and enhance designers’understanding of the performance of the barrier.

The research initiated aims to tackle exactly this aspectof road barrier design by validating numerical appr-oaches and codes with precision testing at the LDTF.Experiments are in preparation, which will reliably rep-resent the main parameters of full-scale tests. A bigpart of the project is supported by the Enterprise DG.Within this framework, there is close collaboration withthe Spanish Technological Centre LABEIN in the areaof modelling and numerical simulations with FiniteElement techniques, and with the French L.I.E.R. (Lab-oratoire d’essais Inrets Equipments de la Route) forconducting some standard full-scale tests, necessaryfor assessing the equivalence of the two approaches.Two producers (one French and one Italian) of deform-able road barriers have been contacted for the pro-curement of the specimens.

The actual tests at the LDTF have experienced somedelay. Two barrier manufacturers have delivered thenecessary testing material and the necessary adapta-tions of the machine are in progress. The impactorshave been constructed and the small specimens forthe mechanical characterisation of the barrier materialunder impact conditions have been fabricated. The firstresults produced by the two partners are currently beingassessed: LIER has performed the two tests with thebarriers from the French manufacturer and LABEIN, usingthe code ABAQUS/Explicit, has attempted to imple-ment a model for the LDTF tests and some preliminarynumerical simulations have been run.

92

13.7 Experimental force-shortening curve characteristic of

the crash-worthiness behaviour of the front car structure.

13.8 Numerical simulation and precision crash testing

results concerning the force transmitted to the right

side-member under floor during the first instants

of impact.

13.9 Numerical simulation and precision crash testing

results concerning the force transmitted to the right half

of the tunnel during the first instants of impact.

Force/displacement at incident bar end

impactor displacement (mm)

Forc

e at

inci

dent

bar

end

(N)

16

14

12

10

8

6

4

2

0

x 105

0 10 20 30 40 50

Right side-member under floor signals

time (sec)

Forc

e (N

)

2000

0

-2000

-4000

-6000

-8000

-10000

-12000

-14000

-16000

x 10-31 2 3 3.5 40 0.5 1.5 2.5 4.5

simulationcrash test

Right side-member under floor signals

time (sec)

Forc

e (N

)

2000

0

-2000

-4000

-6000

-8000

-10000

-12000

-14000

-16000

x 10-31 2 3 3.5 40 0.5 1.5 2.5 4.5

simulationcrash test


13.2.3 Pedestrian Safety

The European Commission has been preparing a pro-posal concerning new technical requirements for carswith the aim of reducing the number of fatalities andthe seriousness of injuries in accidents involving pe-destrians. The major contribution to this effort hasbeen provided by the EEVC (European Enhanced Vehi-cle-Safety Committee), which has drafted a document,where four basic component tests are recommendedfor assessing the pedestrian-friendliness of a car. Theseare: child head-form against bonnet, adult head-formagainst bonnet, lower leg-form against bumper, andupper leg-form against bonnet leading edge. However,the ACEA (Association des Constructeurs Européensd’Automobiles) has raised considerable doubts aboutthe validity of parameter values in these tests, and hasinstead counter-proposed others. In the absence ofagreement between the two proposals, the EnterpriseDG has requested the expertise of ISIS on transient dy-namics.

Despite the tight time constraints, the JRC/ISIS evalu-ation team reviewed and assessed a lot of material col-lected from the relevant literature provided by EEVCand ACEA. In the report prepared [13.10], test condi-tions and reference values of parameters have beenproposed. Particular attention and priority has beengiven to the head-form against bonnet test. Openquestions have been pointed out and the approach ofa mild introduction for these tests is being advocated.

References

13.10 Géradin, M., Halleux, J-P., Solomos, G.,Papadopoulos, I. – Second Comments on a PossibleNormative Action on Pedestrian Safety – TechnicalNote No.I.00.144, Ispra, December 2000.

94

OVERVIEWB. Worth, H. WeisshaüplAt present, some 30% of the total electricity generatedin the EU and Candidate Countries is generated by nuc-lear power plants. With the pressing need to meet theKyoto targets on reduced atmospheric emissions, itmust be expected that over the next 10 to 20 yearsthere will be little or no reduction in nuclear energyproduction. Indeed, considerable international discus-sion focuses on the next generation of nuclear reactorswith enhanced performance and safety features. Forthe European Commission, therefore, nuclear safetywill remain a political priority issue, particularly inview of the enlargement process, the assurance ofadequate safety levels in ageing nuclear installations,and the licensing of advanced and innovative new reac-tor systems.

A recent IAEA (International Atomic Energy Agency)report1 on nuclear safety declares that “Nuclear powerplant safety requires a continuing quest for excellence.All individuals concerned need constantly to be alertto opportunities to reduce risks to the lowest practica-ble level. The quest, however, is most likely to be fruit-ful if it is based on an understanding of the underlyingobjectives and principles of nuclear safety and the wayin which its aspects are interrelated”. This statementechoes the belief that safety in general can be en-hanced most effectively through an improved under-standing of the physicochemical processes involved –a tacit endorsement of the need for continued advan-cements in nuclear science.

Incidents, and especially accidents, within any of theexisting nuclear installations could lead, if not properlymitigated, to serious consequences for the plant or,worse, for the public and the environment. The contin-ued availability of nuclear energy as a secure and relia-ble resource will also impact upon Europe’s economicand competitive well-being. Therefore common safetypractices, rooted in a comprehensive safety cultureoperating at all levels, are highly desirable ingredientsof any communal EU policies on nuclear safety.

Concerns have been expressed recently by the Davi-gnon panel-of-experts over the possibility of dwin-dling competencies in the field of nuclear technologywithin Europe. In recognition of these concerns, theJRC is striving to reinforce its own activities in nuclearsafety and to seek counter-measures to maintain ahigh level of expertise in nuclear technology and reac-tor safety. The JRC is therefore emphasizing the impor-tance of nuclear safety by directing policy-relatedactivities along three lines. First, the integration of theCandidate Countries (and the New Independent States)must progress towards a common safety level and

safety culture. Secondly, to provide assurance that thehigh safety levels of Western plants will be main-tained, particularly in relation to the ageing of plantcomponents and nuclear fuels at high burn-up rates,and to further improve and validate accident manage-ment procedures. Thirdly, provision of the neededcompetence for new innovative reactor concepts (withthe potential to dispose of plutonium) must be main-tained to keep the nuclear energy option open.

The approach to nuclear safety in the Candidate Coun-tries is subtly different to that of the West and cannotdraw on equally broad safety investigations and research.A transfer of the achievements and know-how to thesecountries, covering all key safety issues, is thereforean essential step towards improving the safety of theirplants and to promoting their integration within a com-mon safety level and safety culture. Indeed, it is seen asa prerequisite to their joining the EU. To achieve theseaims, the JRC, with its long-standing involvement in avariety of nuclear safety-related activities, will contin-ue to give direct and impartial support to the Commis-sion Services responsible for nuclear safety and civilprotection.

The main focus of the unit is on the development ofmethods, interpretation of experiments, and code vali-dation for severe accidents, covering important topicssuch as melt-coolant interactions, structural integrity,hydrogen risk, and source term (fission products). Theunit also develops innovative models for plant thermal-hydraulics. A growing activity is the evaluation of thesafety of new reactor designs, which complements themore general activity on harmonisation of safety appr-oaches and safety assessment methodology. Dissemi-nation of data and collaboration, particularly withenlargement countries, are highlighted in the JRC’spolicy support role. Finally, the unit is a leader in ISISin modelling and technological innovation, and anumber of spin-off activities have been supported in2000, both via exploratory research and competitivefunding.

1 IAEA report; Basic Safety Principles of Nuclear Power

Plants, 75-INSAG-3 Rev.1, INSAG-12, IAEA, Oct. 1999.

Nuclear Safety14

14.1 SEVERE ACCIDENTS

A. Annunziato, C. AddabboThe research activities carried out in the accident sim-ulation and analysis sector were mainly aimed at theevaluation of the test results obtained in the per-formed FARO/KROTOS experimental programmes andat the further development/assessment of the JRCmelt-coolant-interaction safety analysis code COMETA.High priority has been given to the development andimplementation of the STRESA web-based informaticsplatform for the management of the FARO/KROTOSand LOBI experimental databases. Collaborative activ-ities with Central Eastern European Countries (CEEC)and the New Independent States (NIS) research organ-isations have been promoted within the provisions ofthe EU enlargement agenda in the field of reactor safetyresearch. In line with the rationales of the prospectedEuropean Research Area, the CERTA thematic network(see section 14.5.1) has been established under the ECNuclear Safety FP-5 assembling 10 major European re-search organisations to consolidate and archive theexperimental databases acquired in European reactorsafety integral system effect thermal-hydraulic testfacilities.

14.1.1 Melt Coolant Interaction

A. Annunziato, C. AddabboA comprehensive research programme dedicated tothe identification and/or verification of melt-coolant-interaction phenomenologies in water coolant reactorssevere accident conditions has been conducted at JRCIspra in the framework of the Euratom reactor safetyresearch programme. The overall research activitieshave included 1) design, construction and operation ofthe FARO and KROTOS test facilities to investigatemelt-coolant-interaction as well as melt spreadingphenomenologies under realistic melt composition(UO2-ZrO2-Zr) and 2) development and application ofthe COMETA (Core Melt Thermal-Hydraulic Analysis)code for test design and pre- as well as post-test ana-lysis of the results.

The progression of severe accidents in water-cooledreactors is characterised by complex multi-phase,multi-component processes which include the interac-tion of a variety of reactor materials under a widerange of parametric conditions. Relevant phenomenawith potential safety implications comprise core deg-radation and melt-down, melt relocation into the reactorpressure vessel lower plenum and potential interac-tion with the residual coolant which could lead toquenching or to energetic escalation with a violentheat transfer from the melt to the coolant. This in turncould results in rapid vaporization with conversion ofthermal energy into disruptive mechanical loads.


14.1 Web-based information platform STRESA.

View films and images

Objectives of the STRESA Database

Large scale experimental programmes have been executed at JRC-Ispra duringthe past two-decades to support development and assessment of system codesused in reactor safety analysis.

Stresa has been developed to provide a web-based informatic platform for thepreservation of the acquired experimental data bases and the maintenance ofthe supporting information and documentation.

Experimental data bases currently included in Stresa-JRC:

• LOBI MOD1 and MOD2: Loss of Coolant Accidents and Transients in PWRs• FARO: Melt Coolant Interaction and Quenching• KROTOS: Steam Explosion

Display and printdocuments on-line

Plot single or multipledata signals

Study designdrawings

Download and analyse datawith post-processing

programmes (e.g. WinGraf)

Retrieve relevant documentsor tests using the

search engine

96

Following the completion of the FARO/KROTOS experi-mental programmes, analytical activities have been fo-cused on the evaluation and documentation of rele-vant test results with inclusion of supporting post-testprediction calculation performed with the COMETA(Core Melt Thermal-Hydraulic Analysis) safety analysiscode [14.1-14.2].

Studies on scaling aspects relevant to the applicationof the COMETA code to a full size reactor geometryhave been conducted in the framework of an universitydegree dissertation grant through the application ofthe code to the analysis of the TMI accident sequences[14.3]. A synopsis of the COMETA predictive capabili-ties benchmarked against FARO test data have beenreported in [14.4].

In order to ensure storage and retrieval of the experi-mental data acquired in the execution of the FARO/KROTOS as well as of the LOBI experimental pro-grammes, the web-based informatic platform STRESA(Storage of Reactor Safety Analysis Data) has been de-veloped to facilitate dissemination and access to thedata within and outside the European Union (Figure 14.1).

14.1.2 Structural Integrity of Reactor Vessel in

Severe Accidents

G. SolomosThe problems of the impact of a slug of molten coriumand debris on the upper reactor vessel head as well asthe dynamic pressure loading in the lower part of thereactor vessel due to steam explosion are of concern.Reliable modelling and simulations require the knowl-edge of the material deformation and failure processes.Strain rate and temperature must be correctly includedand size effects should be examined in order to assurethe transferability of the small specimen results to thereal structure.

ISIS has carried out dynamic tests for material charac-terisation, employing principally the Large DynamicTest Facility (LDTF). Uniaxial and biaxial tension testshave been performed on the materials used for thepressure vessel and its internal structures. These in-clude: ferritic steel 20MnMoNi55 (vessel head), auste-nitic steel X6CrNiNb1810 (Upper Internal Structure),ferritic steel 26NiCr Mo146 (bolting). Geometricallysimilar smooth cylindrical tension specimens of diam-eters 3 mm, 9 mm and 30 mm have been tested atroom and higher temperatures (400˚C-600˚C), and atstrain rates ranging from quasi-static (10-3/sec) to dy-namic (200/sec) conditions. Notched cylindrical tensionspecimens of minimum diameter 2.4 mm, 7.2 mm and2.4 mm have also been tested under the same condi-tions (Figure 14.2). Strain rate and temperature effectshave been adequately assessed through the stress-strain diagrams obtained [14.5]. The size effect issuehas demonstrated to be more sensitive. For the smoothspecimens it appeared to be minor with respect to re-sistances and concentrated mainly to local deforma-tion parameters near fracture, such as the area reduc-tion, the meridional radius of curvature etc. [14.6]. Thetests with the notched specimens, where strain gradi-ent effects and local phenomena are more accentuat-

References

14.1 Annunziato A. et al. – Quick Look Report of FAROTest L-33 -TN.I.00.111

14.2 Annunziato A., Franceschini A., Addabbo C.– TMI Simulation by RELAP5/SCDAP and COMETACodes – ICONE-8 Conference, Baltimore, MD/USA,April 2-6, 2000

14.3 Pla P., Annunziato A., Addabbo C. – Simulacionmediante el codigo de calculo COMETA de losexperimentos de interaccion combustibile fundido-refrigerante realizadios en la instalacion FARO –Reunion Annual de la Sociedad Nuclear Española,Leon-Spain; October, 2000.

14.4 Technical Exchange Agreement between EC-JRC andthe Electrogorsk Research and Engineering Centre(EREC); Electrogorsk, Moscow Region, RussianFederation.

14.2 The ø2.4mm, ø7.2mm and ø24mm nuclear steel

notched specimens used in the investigation for

studying possible size effects.

Ferritic steel 20MnMoNi55

Nom

inal

Str

engt

h F m

(MPa

)

1100

1000

900

800

700

6003

Minimun specimen diameter (mn)0 6 9 12 15 18 21 24 27 30

Notched specimens = 2.4mm, 7.2mm and 24mm

R.T. dynamic 200/s

R.T. static 0.001/s

T = 400˚C, static 0.001/s

T = 400˚C, dynamic 200/s

14.3 Trends of the nominal strength versus size curves for

several strain rate and temperature conditions.

ed, tend to confirm better these size effect trends. Fur-ther, for the notched specimens the nominal strength(maximum force over minimum initial cross-sectionalarea) seems to have some size influence (Figure 14.3).Possible positional effects are examined for explain-ing this behaviour [14.7].

The above activity has been carried out within theframework of two EU projects in the Nuclear FissionSafety/Reactor Safety programme. They are: (a) RPVSA“Behaviour of the Reactor Pressure Vessel under Me-chanical and Thermal Loadings caused by Core Melt-Down and Steam Explosion Accidents”, and (b) REVISA“Reactor Vessel Integrity in Severe Accidents”. A thirdproject is currently in progress: LISSAC “Limit Strainsfor Severe Accident Conditions”. The objective is tostudy whether larger structural deformations shouldbe allowed for severe accidents. Thus, failure strainsof essential reactor components are investigated. Theexperimental work includes testing under differenttemperatures, and static and dynamic loading condi-tions for a wide range of specimen sizes.

14.1.3 Hydrogen Risk

T. Huld, H. WilkeningOne of the potential hazards during a severe accidentin a nuclear reactor is the formation of hydrogen fromthe interaction of the hot core melt with the coolingwater. An ignition of the hydrogen cloud may trigger asevere explosion that could damage the reactor con-tainment. Due to the large-scale complex phenomenainvolved, studies of these explosions necessarily involvenumerical simulations. For this task, the REACFLOWcode has been under development for several years inISIS, and has been applied successfully to a number ofsmall-to-medium scale problems relevant to attaininga better understanding of the underlying phenomena.

During 2000 a number of new developments and im-provements have been implemented in REACFLOW,including large-eddy-simulation (LES) turbulence mod-elling, improved diagnostics, and completion of the

incompressible 3D solver with grid adaptation. Thiswill extend the capabilities of the code to low-speedflows with possible application for hydrogen distribu-tion problems.

Work started on a new shared-cost-action project HYCOM,in collaboration with the Forschungszentrum Karlsruheand the Kurchatov Institute, Moscow, and others. Theaim of this project is to assess the capabilities of hy-drogen combustion codes to predict the consequencesof hydrogen combustion events on a large scale, in-cluding a full-size nuclear reactor enclosure. For thisproject a number of blind predictive calculations havebeen carried out for comparison with experiments per-formed in the large-scale RUT explosion facility in Russia.

Work on another shared-cost-action project EIHP, tostudy the safety aspects of using hydrogen for vehiclepropulsion, was brought to a successful close. Thiswork involved a collaboration with a number of part-ners from industry (BMW, Renault, Air Liquide, Volvo,Messer Griesheim and others) and research institu-tions (Demokritos National Laboratory, Greece). A fol-low-up to this project, EIHP-2, has been proposed andaccepted during 2000. The scope of this project will beto study a broader class of safety issues related to theuse of hydrogen for vehicles, in particular regarding thesafety of fuelling stations and distribution networks.

14.1.4 Source Term

14.1.4.1 Phébus Experimental Programme

R. ZeyenUnder the Phébus international in-pile Fission Product(FP) release programme, five in-pile bundle degrada-tion experiments have been performed with specialemphasis on fission product releases from degradedcorium, transport and deposition in the coolant circuitand containment. FP chemistry, particularly iodine inthe isolated containment is then a primary objective ofthis global severe accident research programme.While 1999 was marked by the successful performanceof the debris bed test FPT4, with a large amount of dataon the melt processes of a highly irradiated fuel mix-ture debris bed (UO2 + ZrO2), this year was speciallyhighlighted by the running of the full bundle test FPT2test on 12 October, 2000.

FPT2 is the third bundle test with the main characteris-tic that the fuel degradation process will be carried outin “steam starved” or partially-hydriding conditions.Boric acid is introduced into the steam cooling to par-ticipate in complicated chemical reactions with fissionproducts and structural materials from the fuel bundle.


References

14.5 Solomos, G. et al. – Experimental investigation ofStrain Rate, Temperature and Size Effects in NuclearSteels – Technical Note No.I.00.76, JRC Ispra, June2000.

14.6 Malmberg, T., Krompholz, K., Solomos, G., Aifantis, E.C.– Investigation on size effects in ferritic and austeniticmaterials – Proceedings SMiRT-15, Seoul, Korea, 15-20August 1999.

14.7 Solomos, G. et al. – Experimental investigation ofSize effects in Nuclear Steels using Notched TensileSpecimens – Technical Note No.I.00.136, JRC Ispra,November 2000.

98

Several preliminary conclusions can already be drawnfrom this test:

• The fuel degradation happened at higher tempera-tures and higher reactor power.

• An extended full hydrogen release period could benotice (as expected).

• A substantial amount of gaseous iodine from the cir-cuit arrived into the containment, less iodine retentionin the sump due to the absence of silver from thecontrol rod material in the melting process.

• High aerosol releases from the fuel into the contain-ment followed by fast settling.

Fuel and test section tomography reveals, throughdensity shadings, the different relocation zones anddegradation mechanisms and, from post-test-exami-nations, local corium melting temperatures.

FP samplers are now being extracted remotely fromthe circuit for subsequent analysis in a number of spe-cialised European hot laboratories. The Institute forTransuranium Elements (ITU) Karlsruhe is the mostwell-equipped and well-trained laboratory in this field.The final data report is not expected before 2002. Thefinal interpretation reports will come out 6 or 7 yearsafter the test performance.

Hydrogen recombiners were also tested against po-tential poisoning by real fission products. A holdingdevice was designed to introduce, during a given timeinterval, coupons from all major world recombinermanufacturers. Unfortunately this device, after severaltrials, could not be introduced into the aggressive at-mosphere of the simulated containment, such that wewill have to wait for several years before this missingrecombiner qualification can be performed.

The next test will be FPT3, now called the boron car-bide test and scheduled for March 2003. Here the Ag,In, Cd control rod will be replaced by the B4C, moretypical for BWRs but also for more modern FrenchPWRs. Finally, not before 2005, the FPT5 test will prob-ably be a bundle degradation test including an air ingressphase.

ISIS’s Nuclear Safety unit is, as before, present as a co-leading agent in all executive working groups andsteering committee meetings, as well as by its presencein day-to-day management of a number of technical andorganisational matters in this important programme.International contributions to the Phébus programmefrom different countries are directly managed by ISISscientific personnel. The Phébus Steering Committeein 2000 decided the conditions for test FPT2, includingthe testing of recombiner coupons, and approved apreliminary specification of the next test, FPT3. JRCexperts have continued to support the Research DG byreviewing the experimental reports produced by IPSN,a considerable task in 2000. The Preliminary Report oftest FPT4, and the Quick Look Report of test FPT2 haveboth been reviewed by the JRC. For the Final Report oftest FPT1 additional reviewing effort was enlisted fromvolunteer experts from AEA Technology, FZK, CIEMATand Chalmers University (Sweden), and a large numberof pertinent comments were collected, helping to pro-duce a much improved final version of the report.

14.4 The international Phébus FP Programme is situated at

the CEA Cadarache in southern France, co-financed by

the European Commission (Research DG), EdF and a

number of overseas partners like USNRC, NUPEC and JAERI

from Japan, KAERI from Korea, COG from Canada. More

recently HSK/PSI from Switzerland have joined the pro-

gramme co-financing by direct and in-kind contributions.

The general objectives of this severe accident research

programme are on fission product release and transport

studies from degraded and molten nuclear fuel at medium

and high burn-up,

The study comprises the release ratios and kinetics of

high-, medium and low volatile fission products and

structural materials.

The FP chemistry in the containment is another one of

the primary objectives.

Secondary objectives include testing of hydrogen cata-

lytic recombiners against real fission product poisoning.

The Phébus Reactor at CEA Cadarache


14.1.4.2 Phébus Network

A.V. JonesISIS continues to contribute to the interpretation ofthe Phébus FP test results, the preparation of futuretests, and the validation of LWR severe accident compu-ter codes and models against Phebus data. Phenomenaconcerned are core degradation and fission productrelease and source term issues of fission producttransport in the reactor circuit, aerosol removal in thecontainment, and both short-term and long-term fissionproduct chemistry. It also promotes European and in-ternational co-operation on these themes through thePhébus “scientific analysis working group” (SAWG)and its interpretation focuses on degradation/release,aerosols and chemistry. These groups, which not onlyserve the Phébus project but also act as a focus for severeaccident research across Europe, met twice in 2000.The results from related shared-cost action projectswhere the JRC is a participant [14.8] are also factoredinto the activities, known generically as PhébusNet.One product in 2000 has been a status report on corequenching for the OECD, written in collaboration withGRS [14.9]. A well-attended Phébus Workshop, held inMarseilles (jointly with IPSN) in February 2000, includedJRC papers on the state of understanding of bundledegradation, hydrogen generation and fission productbehaviour in the Phébus FP tests performed so far[[14.10-14.12]. Also presented was a paper on the appli-cation of Phébus results to plant analysis in Europe, theoutcome of a survey of 14 safety authorities, designersand technical support organisations in 7 Europeancountries [14.13]. Respondents identified several spe-cific applications of Phébus data, including code vali-dation and well-identified safety issues. They wereunanimous in expressing appreciation for the way inwhich Phébus supports a holistic view of severe accidentphenomena, and for the cooperative spirit engenderedby the team approach adopted to test interpretationand analysis. French take-up of Phébus results was thesubject of a separate paper.

Within PhébusNet, ISIS has made and applied somesignificant model developments in 2000. Regardingaerosols, progress has been made in the theory of parti-cle formation [14.14], and analytical or semi-analyticalapproaches have been developed to heat transfer andparticle deposition in tubes [14.15-14.18] which improvepredictions for fission product retention in the reactorsteam generator compared with current codes. In ad-dition, an advanced random-walk simulation studywas made of thermophoretic deposition in turbulentboundary layers [14.19]. Concerning aerosol resuspen-sion, the JRC’s CAESAR code was assessed againstSTORM experiments [14.20], and a more general surveyhas been made of kinetic resuspension models [14.21].In a collaboration with GRS and with VEIKI (Hungary) avisiting scientist is validating the latest version of theCOCOSYS code against containment thermal-hydraulicsand aerosols information from the Phébus test FPT0.

Chemistry strongly affects the mobility of fission pro-ducts and is subject to considerable uncertainty inPhébus. A collaboration with IPSN has continued onthe kinetic model for circuit chemistry, described inthe 1999 Annual Report. Recently developed modelsfor silver-iodine interactions and for organic iodide for-mation have been applied in a prediction of contain-ment behaviour in the very recent test FPT2, where thedegradation and release took place in steam-poor con-ditions [14.22]. Low iodine concentrations in the con-tainment atmosphere were predicted, a result whichseems to be confirmed by on-line measurements. Furtherwork on iodine chemistry has been undertaken underthe shared-cost action project ICHEMM.

Early in 2001 the Final Interpretation Report of testFPT0 should be completed, and the Intermediate Inter-pretation Report of FPT1 is scheduled for the latter halfof the year. Precalculations for the ground-breakingtest FPT3 should also begin.

14.5 The FPT2 ultrasonic thermometers performed extremely

well and indicated temperatures higher than 2500°C.

Also the 100% hydrogen phase, lasting about 18

minutes can be well recognised.

14.6 FPT0 steam-generator deposition with new model versus

experimental data and predictions from other codes.

FPT2 test:Temperatures and H2 generation

TUS2.6 (630) TUS2.5 (550)TUS2.4 (470) TUS2.3 (390)TUS2.2 (310) TUS2.1 (250)SDHY701

Temperatures (˚C)

Hyd

roge

n ge

nera

tion

(%)

2900

2700

2500

2300

2100

1900

1700

1500

1300

1100

900

700

5007000 8000 9000 10000 11000 12000 13000 14000 15000 16000 17000 18000 19000 20000

50

40

30

20

10

00.011 E-3 0.1 1

2-D model 3 g s-1

1.5 g s-1

experimental

Batchelor & Shen

1-D model

RAFT (VICTORIA, SOPHAEROS)

Particle radius µm

% R

ete

nti

on

100

14.1.4.3 PHEBEN2A.V. Jones

The shared-cost action project PHEBEN2 stimulatesthe application of both detailed and integral (system-level) codes in use in the EU to the analysis of PhébusFP results, taking advantage of the existing structureof PhébusNet. Its other objectives are to validate thecodes against the Phébus data and to develop and applyassessment criteria for integral codes as used in theanalysis of real plants. In the first year of the projectthe 13 partners have contributed to the FPT0 FinalInterpretation Report and performed analyses of FPT1data using both types of code, and a compendium ofsummary descriptions of the integral codes in use bythe partners, ASTEC, ECART, MAAP and MELCOR, hasbeen drafted. In 2001 integral code assessment willplay a larger role in the project’s activities. Synergywith the planned OECD ISP 46 on the integral analysisof Phébus test FPT1 is expected.

14.1.4.4 EVITA

A.V. Jones, K. MuellerIn the frame of the shared-cost action project EVITA,the code ASTEC V0.3 developed by IPSN and GRS wasinstalled on a SUN as well as on an ALPHA VX worksta-tion. The aim of this ongoing code development is toprovide the end-users like utilities, vendors and licens-ing authorities with a fairly well validated EuropeanIntegral Code for the simulation of severe accident se-quences. Different reactor test cases were recalculatedin order to check the installation and to obtain firstexperience in code handling. To apply the implement-ed fission transport module SOPHAEROS V2.0 to thePHEBUS-FP experiments, an input deck for the steamgenerator tube (SGT) was generated. The first prelimi-nary results indicate that the retention in the SGT wasgreater as a previous calculation showed. Also in thiscalculation a significant overestimation of the meas-ured data was obtained, i. e. the model of thermo-phoretic deposition which is the dominant process hasto be improved. Therefore the development of a sepa-rate particle tracking model was started. For code bench-marking using the SGT the fission product release andtransport code VICTORIA 2.0 was installed. Future bench-mark calculations will also be performed with thesecodes using selected STORM experiments.

14.1.4.5 ICHEMM

Y. Drossinos, E. KrausmannThe objective of the shared-cost action project ICHEMM“Iodine Chemistry and Mitigation Mechanisms” is tostudy, both experimentally and theoretically, aspectsof iodine chemistry that are currently poorly under-stood, especially for BWR conditions. In particular,some processes that lead to the destruction of volatileforms of iodine are not well quantified. An improved

References

14.8 The shared-cost actions ASTERISM II, COLOSS,PHEBEN2, EVITA, ICHEMM are described separately.

14.9 Haste, T.J., Trambauer, K. – Degraded Core Quench:Summary of Progress 1996-1999 – ReportOECD/NEA/CSNI/R(99)23 (February 2000).

14.10 Shepherd, I.M. – 4th Phebus FP Technical Seminar –Marseilles, 20-22, March 2000.

14.11 Jones, A.V., Kissane, M. – State of Understanding ofFission Product Transport in the Circuit and of Aero-sol Behaviour in the Containment of Phebus FP – 4th

Phebus FP Technical Seminar. Marseilles, March20-22, 2000; also in European Aerosols Conference2000, Dublin, September 2000.

14.12 Drossinos, J. et al. – 4th Phebus FP TechnicalSeminar – Marseilles, March 20-22, 2000.

14.13 Weisshaeupl, H., Jones, A.V., Zeyen, R. – Use ofPhebus Results in Other European Countries –4th

Phebus FP Technical Seminar. Marseilles, March20-22, 2000.

14.14 Drossinos, Y., Kevrekidis, G., Lazaridis, M.,Georgopoulos, P.G. – Translational Invariance in theTheory of Nucleation – 15th Int. Conf on Nucleationand Atmospheric Aerosols, Rolla, Missouri (August2000).

14.15 Ezquerra Larrode, F., Housiadas, C., Drossinos, Y– Slip-flow Heat Transfer in Circular Tubes – Int. J.Heat and Mass Transfer 43, 2669-2680, June 2000.

14.16 Housiadas, C., Ezquerra Larrode, F., Drossinos, Y.– Convective Diffusion in a Tube with Non-uniformInlet Conditions – J. Aerosol Sci. 31, 995-968,August 2000.

14.17 Ezquerra Larrode, F., Housiadas, C., Drossinos, Y.– The effect of coupled heat and mass transfer on thesaturation ratio in laminar tube flow – J. Aerosol Sci.31, S930-S931, September 2000.

14.18 Lazaridis, M., Drossinos, Y., Spiraki, A., Georgiadou,Gekas,V. – Sensitivity calculations on new particle for-mation parametrization using a modal aerosol dynamicsmodel – J. Aerosol Sci. 31, S442-S443, September 2000.

14.19 Kroeger, C., Drossinos, Y. – A Random-walk Simula-tion of Thermophoretic Particle Deposition in aTurbulent Boundary Layer – Int. J. Multiphase Flow26,1325-1368. August 2000.

14.20Hontanon, E., De los Reyes, A., Capitao, J. – The CAE-SAR Code for Aerosol Resuspension in TurbulentPipe Flows - Assessment against the STORM Experi-ments – J. Aerosol Sci. 31,1061-1076, March 2000.

14.21 Reeks, M., Hall, D. – Kinetic Models for Particle Re-suspension in Turbulent Flows: Theory and Meas-urement – J. Aerosol Sci. 32,1-31, September 2000.

14.22Krausmann, E. – Phebus FP: An IMPAIR3 Pre-testCalculation of the Phebus FPT2 Test – JRC Tech NoteJRC TN I.00.112 (September 2000).

14.7 IMPAIR3 predictions for gaseous iodine in FPT2.

10-9

10-10

10-11

10-12

10-13

10-14

10-15

10-16

10-17

0 100000 200000 300000

I gas

[mol

/l]

t [s]

I2

Itot (I2 + RI)

RI


knowledge of these destruction rates will allow theirimportance to be assessed in terms of natural mitigationprocesses and accident management interventions.Moreover, the effect of certain materials and conditionsspecific to BWR systems on iodine behaviour are notknown. During the first year, a state-of-the-art reporton “Iodine chemistry and related mitigation mecha-nisms” [14.23] was completed. Furthermore, iodinebehaviour in the Phebus-FPT0 experiment was calcu-lated with an improved IMPAIR code; new models forheterogeneous organic iodide formation were includ-ed. The results showed better agreement with experi-mental data.

14.2 THERMAL HYDRAULICS

H. Städtke, B. Worth, G. FranchelloModel development in fluid dynamics continues to bean active research topic within the ISIS nuclear safetyarea, and computational fluid dynamics (CFD) is widelyused for the simulation of generic thermal-hydraulicflow processes. At the JRC, a continuous effort in theevolution of basic two-phase flow theory has resultedin the development of new and innovative methodolo-gies, providing a mathematical basis for future codedevelopment in this field [14.24-14.26]. These methodstake advantage of high-resolution “up-winding” tech-niques now widely used in aerodynamics and compressi-ble gas dynamics. They are developed specifically tocapture the fine detail of discontinuous flow processes,such as shock wave propagation or, as in the case ofnuclear reactors, the dynamic motion of two-phasemixture levels formed in the primary circuit duringloss-of-coolant accidents.

The “Advanced Two-Phase Flow” (ATPF) pilot code beingdeveloped within ISIS provides one such approach forthe numerical simulation of transient multi-dimensionaltwo-phase flow processes. This work is supported partlythrough EC-funded competitive shared-cost activities.The ATPF code is based on a fully-hyperbolic “two-fluid” model for the simulation of non-equilibriuminhomogeneous two-phase flow using separated con-servation equations for each of the two (coupled)phases. A mathematical approach based on the exist-ence of real eigenvalues (characteristic velocities) anda complete set of independent eigenvectors allows thefluxes to be calculated from the wave propagationcharacteristics (eigenstructure) of the flow. This facili-

tates the implementation of stable “upwind” numericalschemes based on approximate Riemann Solvers, whichcombine signal propagation along characteristic linesto ensure conservation of mass, momentum and energy.A second-order finite volume interpolation techniqueprovides high-resolution simulation of two-phase flowwith very low inherent numerical viscosity. Real fluidviscosity can be included to give a physically-basedsimulation of viscous Navier-Stokes type flow problems.

The ATPF code has been successfully used for study-ing a large number of problems typical of those in thenuclear and process engineering industries. Of particularrelevance to PWR safety is the simulation of steam-watermixture levels in a reactor pressure vessel during de-pressurization (“blowdown”) transients resulting fromloss-of-coolant accidents. These situations can lead torapid core uncovery, dryout of the fuel rod bundle andpossible (but extremely improbable) “melt-down”events. An accurate prediction of in-core temperaturevariation can be made only if the wetted regions of thecore can be described from knowledge of the two-phasemixture levels. Resolving this problem with existing in-tegral systems codes (e.g. RELAP5, CATHARE2, ATHLET)requires relatively fine discretization of the core, lead-ing to very lengthy computation times. Other problemsof interest include the accurate treatment of criticaltwo-phase “choked” flow with steep pressure gradientsand thermal non-equilibrium, fast depressurizationand “flashing”, slow gravity-induced phase separationand the simulation of two-phase flow processes withlow driving heads, such as natural convection processes.

References

14.23 Krausmann E. – A state-of-the-art report on iodinechemistry and related mitigation mechanisms –manuscript (2000).

References

14.24 Städtke H., Franchello G., Worth B. – NumericalSimulation of Gas-Liquid two-phase Flow Based onHyperbolic Flow Equations – AMIF-ESF Workshopon Computing methods for Two-Phase Flow,Aussoise, France, 12-14 January 2000.

14.25 Tuomi I., Barre F., Städtke H., Graf U., Mimouni S.– Advances in Numerical Methods for Two-phaseFlows – OECD/CSNI Workshop on AdvancedThermal-hydraulic and Neutronic Codes,Barcelona, Spain, 12-14 April 2000.

14.26 Städtke H., Franchello G., Worth B.– Numerical Simulation of Two-phase Flows withStiff Source Terms – First. International Conferenceon Computational Fluid Dynamics, Kyoto, Japan,9-14 July 2000.

102

The capabilities of the JRC ATPF code in simulating suchproblems are indicated in the ROSA-1 vessel blowdownproblem (Figure 14.8 and 14.9). In this case, two cylin-drical vessels (treated as pseudo 3-dimensional abouta plane of symmetry) are connected by a horizontal(1-dimensional) pipe in which an orifice is located torestrict the flow. Initially, one vessel is partially filledwith hot saturated water at 10 bar pressure whilst theother is filled with pure saturated steam at 1 bar. Atransient event is initiated by rupturing a bursting diskat the outlet of the high-pressure vessel (Figure 14.8

left), causing rapid boiling with the discharge of two-phase fluid from the tank. Measured and predicted pres-sures, and residual mass inventory, as functions of time,are shown in Figure 14.8 for two different orifice sizes.Figure 14.9 shows the predicted volume fraction forliquid (dark blue) and vapour (light blue) in the left-hand vessel at four different times after initiating thetransient. Some interesting flow patterns, easily iden-tifiable in the prediction, include: rapid swelling of themixture level due to fast evaporation in the liquid pool,transition from single phase vapour to two-phase flowand “choking” in the interconnecting pipe, jet forma-tion in the low pressure vessel and jet impingement atthe right-hand vessel wall (not shown), strong recircu-lating flow patterns, gravity-induced phase separation,liquid collapse and a wave-like “sloshing” of residualliquid pools in the two vessels.

14.2.1 ASTAR

H. Städtke, G. Franchello, B. WorthThe main objectives of the ASTAR (Advanced 3-D Two-Phase Flow Simulation Tool for Application to ReactorSafety) shared-cost action project, which started in2000, are to establish the scientific and mathematicalbasis for a new generation of thermal-hydraulic (TH)codes for reactor safety studies. This activity aims atsubstantially improving the multi-dimensional predic-tion capabilities as well as enhancing those capabilitiesin the present generation of system codes, for safety-relevant phenomena in existing and future innovativeLight Water Reactors. As a competitive activity, the ASTARproject involves partners from seven European organi-sations including the French partners CEA and EdF(Electricité de France), the European Commission, GRS(Germany), Manchester Metropolitan University (UK),the Von Karman Institute for Fluid Dynamics (Belgium)and the Paul Scherrer Institute (Switzerland).

The basis for the development of improved two-phaseflow simulation is the 3-D “two-fluid model” coupledto the transport equations for interfacial area and tur-bulent kinetic energy and dissipation rate, for whichnew experiments are being planned to permit the de-velopment and validation of new physical models. High-resolution numerical schemes with very low artificialdissipation will be further developed and their improvedaccuracy will be assessed by extensive benchmarkingon safety-relevant flow problems, and by comparingwith prediction capabilities of existing TH codes. Thedevelopment and verification of new 3-D TH code mod-ules components are expected to overcome many ofthe deficiencies and limitations of present TH-systemscodes like CATHARE, ATHLET, TRAC or RELAP5. Thesecomponent modules could form the development basisfor a new generation of advanced TH-codes having ahighly modular structure, making them easily adapta-ble to coupling with existing European TH-codes in thenear future.

experiment

prediction

time [s]0 50 100 150

80

60

40

20

0

pres

sure

[bar

] orifice d = 50 mm

orifice d = 70 mm

experiment

prediction

time [s]0 50 100 150

1.5

1.0

0.5

0

mas

s [t

on] orifice d = 50 mm

orifice d = 70 mm

14.8 Predicted and measured pressures and mass inventory in the ROSA-1 Vessel Blowdown tests. Simulated discharge of a steam

and water mixture from a cylindrical pressure vessel via a horizontal outlet pipe, as a function of time for two different orifice

sizes, made with the JRC “ATPF” pilot code.

14.9 Predicted volume fraction and velocity fields of the gas

and liquid phases in the ROSA-1 Vessel Blowdown tests,

for various times after rupturing the bursting disk;

simulation made with the JRC ‘ATPF’ pilot code

d = 0.56morofice

ruturedisk

h = 7.10mh =6.00m

t = 0.0s t -5s t -10s t -20s t -40s t s


14.3 SAFETY OF NEW DESIGNS

H.U. Wider, J. Karlsson, A.V. JonesThe main task of this exploratory research activity wasto gather information on heavy-metal-cooled criticalreactors, now receiving much attention worldwide.ISIS has also contributed to this debate [14.27-14.28].At the ICONE8 conference in Baltimore in April 2000,two sessions were dedicated to Lead Bismuth Eutectic(LBE) – cooled critical reactors. An interesting designpresented by ANL is a 300 MWt pool-type modular re-actor with a secondary loop wherein water is super-heated in the heat exchangers located in the upperpart of the downcomer. This configuration makes thereactor compact and therefore cheaper, but a leak inthe heat exchangers would lead to pressurisation ofthe vessel. The first ANL design featured natural circu-lation but jet pumps have since been added, in serieswith the heat exchangers, to maintain good natural cir-culation flow. Smaller designs by University of Berkeleyfeature cooling of the LBE via a corrugated inner vesselwall by superheated water, all surrounded by a strongexternal vessel. The whole core can be removed like abattery. This design would thus be a very proliferation-safe system. A design from MIT proposes the bubblingof (superheated) water though the riser of the core.The problem of this elegant looking approach is the α-radioactivity of the LBE coolant. At the Generation IVmeeting in Washington (May 2000), key criteria suchas low capital cost (i.e., less than 1000 $/kW installed),demonstrable safety, low waste production and highproliferation resistance were identified as desirabledesign features. An LBE-cooled modular reactor is aGeneration IV candidate in the US.

Minatom of Russia announced, in July, the starting ofthe BREST-300 project. This should lead within 10 yearsto a 300 MWe lead-cooled critical reactor and an inte-grated reprocessing facility for multiple recycling ofthe mixture of plutonium and minor actinides as wellas the water-soluble long-lived fission products. TheBREST-300 reactor is claimed to be cost-competitive,inherently and ecologically safe as well as prolifera-tion-proof since no breeding blankets will be used andthe plutonium will not be separated from the minor ac-tinides. Obninsk in Russia, however, is proposing thesmaller SVBR-75 LBE-cooled reactor based on their80-reactor years experience with submarine reactorsof this type.

The Tokyo Institute of Technology has for a long timeproposed small and transportable super-safe LBE-cooled reactors with fuel lifetimes of 12 years. SouthKorea is proposing the LBE-cooled PEACER reactorthat should meet the Generation IV criteria. Sincethere is no research yet in the EU on this reactor type,ISIS started an initiative for a concerted action consid-

ering the BREST-300 reactor and the different LBEcooleddesigns. The main objectives are to determine whetherto propose an EU co-operation with BREST-300 and todetermine whether this type of reactor could be consid-ered acceptable in the EU. A proposal has been submitt-ed involving experts from Russia (RDIPE and Obninsk),six EU countries, Switzerland, Japan and Korea.

14.4 HARMONISATION

Overview

G. CojazziISIS promotes harmonisation on nuclear safety aspectsby providing technical support to the relevant EuropeanCommission DGs by participating in international ac-tivities such as those of the OECD and IAEA throughnetworking laboratories, by organising benchmark exer-cises and performing focused research on safety criticalissues and risk-assessment methods.

ISIS supports the Nuclear Safety and Civil ProtectionDirectorate of the Environment DG, in the implementa-tion of the two Council Resolutions (C128-75, C172/2-92)dealing with reactor safety. The activity performed withinthe Commission permanent advisory group (the NuclearRegulator Working Group (NRWG)), which networksthe Nuclear Regulatory bodies of EU Member States,have been closely followed. Support in this sensitivearea will continue with the Energy and Transport DG towhich the nuclear safety matters in EU Member Stateshave been transferred during the course of the year.

One of the main concerns among EU citizens andstakeholders is the safety of nuclear power plants ofEastern design. In order to improve the safety of theseinstallations, the Commission has devoted considera-ble resources to technical assistance programmes. Asa spin-off of the competencies acquired in reactorsafety, JRC-ISIS together with JRC-ITU, JRC-IAM andJRC-DISI continued in 2000 to support the technicalimplementation of the Commission’s TACIS and PHAREassistance/co-operation programmes with CIS andCEEC [14.29]. ISIS’s contribution focuses on its areas ofcompetence, namely thermal-hydraulics, design basisand beyond design basis accidents, severe accidents,

References

14.27 Wider H.U., Karlsson J.K., Jones A.V. -– “Lead/Bismuth - Cooled, Thorium Based ADS andCritical Systems” Meet Sceptic’s Criteria –ICENES2000 International Meeting in Petten,The Netherlands, September 2000.

14.28 Wider H.U., Karlsson J.K., Jones A.V. – Aspects ofSevere Accidents in Transmutation Systems – OECDmeeting on Partition and Transmutation in Madrid,December 2000.

104

risk assessment, structural dynamics and integrity.Support has been given to the Common Service for Ex-ternal Relations (now EuropeAid Co-operation Office –AIDCO), for “design safety” projects, in the drafting ofcontract Terms of Reference, in the selection and eval-uation of project offers, in the follow-up and in the reviewof intermediate and final project reports. ISIS has alsoprovided technical assistance, to the Environment DG,in the review and evaluation of project reports relatedto strengthening Eastern Technical Support Organisa-tions (TSO) [14.30-14.31].

14.4.1 Probabilistic Safety Assessment

Probabilistic Safety Assessment (PSA) is one impor-tant area in which ISIS has developed knowledge andcompetence over many years. The subject and relatedmethods are of relevance also with respect to the generaltrend of developing and using risk-informed methodsin support of decision making.

One of the main projects performed dealt with theproblem of “expert judgement”. This involves how jud-gements derived from experts about safety-relevantissues, together with the relative uncertainties, can becollected from experts in a disciplined way, properlydocumented and, if necessary, suitably aggregated inorder to provide the uncertainty quantification of therelevant issues in a PSA study. ISIS designed and co-ordinated an international benchmark exercise in orderto collect information about the use of structured expertjudgement techniques among level-2 PSA researchersand practitioners and to compare methods. The prob-lems considered were fuel coolant interactions andthe event of hydrogen deflagration/detonation in apressurised water reactor system of evolutionary de-sign. In the course of 2000, the analysis and valorisa-tion of the results of the overall benchmark project hasbeen carried out [14.32-14.39].

The development of the JRC-ISIS code ASTRA for relia-bility analysis, based on a binary decision diagramtechnique, has been continued in synergy with otherresearch lines of ISIS and a module for event tree ana-lysis has been added.

An emerging topic in the safety of nuclear powerplants is related to the development and spreading ofdigital technologies and to their use in safety criticaland safety related systems within NPP, leading to theso-called “safety critical software issue”. The subjectis of relevance for both existing installations, due tothe change from analogue to digital control systems,and for new installations which will make full use ofdigital technologies. In the course of the year 2000, aninternational benchmark study has been designed andaccepted for funding as an action of the Research DG

within the Fifth Framework Programme for comparisonof methods for assessing the dependability of safetycritical software. Partners of ISIS include GRS (Germany),STUK (Finland), VTT (Finland), SIEMENS (Germany), andIPSN (France) [14.40].

A workshop on “Technical Harmonisation of Risk-Based Decision-Making” was organized in collabora-tion with the Risk Management and Decision Supportunit of ISIS and was attended by 120 participants. Theworkshop was a first step in launching an internationalinitiative to promote the development of a genericstandard for risk assessment to be applicable to differ-ent domains and industries [14.41-14.42].

References

14.29 Wider H.U., Cojazzi G., Kirchsteiger C., Jones A.V.,Sevini F., Bièth M., Fabbri L., Monjaret J.L.– Supportof the European Commission funded Phare andTacis Nuclear Safety Programmes by the Joint Re-search Centre (JRC) – Fifth International InformationExchange Forum “Safety Analysis for NPP of VVERand RBMK Types”, 16-20 October 2000, Obnisk, RussianFederation.

14.30 Cojazzi G., Fogli D., Price P., (Eds.) with contributionsfrom: Contini S., Graf U., Kirchsteiger C., Jones A.V.,Papadopoulos I., Solomos G., Wider H., Yerkess A.– Support to DG XI in Review of PHARE and TACIS“TSO Project” Reports – Final Report of the Adminis-trative Arrangement N° B7-5200/97/000874,ISIS/DGXI-AA-97/finrep.doc, EC-JRC Technical NoteNo. I.00. 98, July, 2000.

14.31 Cojazzi G. (Ed.) with contributions from: Contini S.,Franchello G., Graf U., Grassini G., Kirchsteiger C.,Jones A.V., Papadopoulos I., Price P., Wider H.,Yerkess A. – Support to DG XI in Review of PHAREand TACIS “TSO Project” Reports – Interim Report ofthe Administrative Arrangement N° 15046-1999-05A1CO ISP, ISIS/DGXI-AA-99/int-rep.doc, EC-JRCTechnical Note No. I.00.97, July 2000.

14.32 Cojazzi G., Delaval M., Grassini G. – BenchmarkExercise on Expert Judgement Techniques in PSAlevel 2, Phase 2 – Assessment of Hydrogen DDTIssue in a Generic Evolutionary Pressurised Reactor-Interviews with Experts, within application ofKEEJAM methodology, EC-JRC-ISIS, Technical NoteNo. I.00.33, February 2000.

14.33 Cojazzi G., Delaval M., Grassini G. – BenchmarkExercise on Expert Judgement Techniques in PSAlevel 2, Phase 2 – Assessment of Hydrogen DDTIssue in a Generic Evolutionary Pressurised Reactor-Interactions with Experts, within application ofKEEJAM methodology, EC-JRC-ISIS, Technical NoteNo. I.00.34, February 2000.

14.34 Cojazzi G., Fogli D., Pulkkinen U. – Benchmark Exer-cise on Expert Judgement Techniques in PSA level 2,Phase 1: Summary of Self Assessments and Peer Re-views of Structured Expert Judgement Approaches –EC-JRC-ISIS Technical Note, I.00.119, September 2000.

14.35 Cojazzi G., Fogli D., Grassini G. – Analysis of theBE-EJTs results for the prediction of a Fuel CoolantInteraction experiment – PSAM 5 InternationalConference on Probabilistic Safety Assessmentand Management, Osaka, Japan, 27 November – 1December, 2000.

14.5 DISSEMINATION AND COLLABORATION

14.5.1 CERTA

C. Addabbo, A. AnnunziatoThe safety evaluation of existing and, in perspective,of evolutionary or innovative reactor concepts, is gen-erally supported by a wide spectrum of experimentaland analytical activities. These are aimed at: 1) theacquisition of representative experimental databasesin integral system effect and/or separate effect testfacilities, and 2) the development of computer codes toprovide realistic predictions of system and/or compo-nent response and the efficiency of related emergencysafety systems under a variety of postulated accidentand transient conditions.

In the past few decades, a considerable amount ofresources has been devoted at the international levelto the establishment and conduction of experimentalprogrammes relevant to water-cooled reactor safetyanalysis. The extent to which the acquired experimen-tal databases are preserved and can be eventuallyaccessed and retrieved is an issue often debated in thenuclear community. In addition to the progressive lossof skilled human resources with lack of adequate re-placement under the current constraints, a compound-ing problem is given by the continuous advancementof computer hardware and software technologies. Thisis making several of the traditional storage methodsrapidly obsolete and access/retrieval of data is practi-cally impaired.

The programmatic objective of the European thematicnetwork CERTA (Consolidation of the Integral SystemExperimental Data Bases for Reactor Thermal-HydraulicSafety Analysis) is thus aimed at providing a consoli-dated framework for the preservation of the integralsystem experimental databases acquired in the contextof the water reactor safety research programmes carriedout by European industrial and institutional researchorganisations.

JRC contributes to the CERTA network with the LOBI ex-perimental database which has been acquired and de-veloped at the JRC Ispra in the framework of previousreactor safety research programmes and under con-tractual agreement with the former BMFT, Germany.The LOBI database comprises 70 experiments coveringmainly thermal-hydraulic phenomenologies relevantto Design Basis Accidents (DBA) and Transients as wellas emergency operating procedures and accident man-agement strategies. The LOBI test facility consisted ina scaled full-power, full-pressure integral system testfacility representing a 1:700 scale model of a 1300 MWePWR of Siemens-KWU design. As structured, CERTA in-cludes experimental programs relevant to reactors inoperation in the EU member countries (i.e., PWRs and


References

14.36 Cojazzi G., Fogli D., Grassini G., Coe I. M. – Bench-marking structured expert judgement methodolo-gies for the assessment of hydrogen combustion ina generic evolutionary PWR – PSAM 5 InternationalConference on Probabilistic Safety Assessment andManagement, Osaka, Japan, 27 November - 1December 2000.

14.37 Cojazzi G., Fogli D., Grassini G. – (2000) BenchmarkExercise on expert judgement techniques in PSAlevel 2, Phase I: Summary and Evaluation of the Un-structured and Structured Expert Judgment Results– EC-JRC-ISIS, to be published as EURATOM report.

14.38 Cojazzi G., Fogli D. – (2000) Benchmark Exercise onexpert judgement techniques in PSA level 2, ExtendedFinal Report, EC-JRC-ISIS, to be published asEURATOM report.

14.39 Cojazzi, G., Fogli, D., Grassini, G., De Gelder, P.,Gryffroy, D., Bolado, R., Hofer, E., Virolainen, R.,Coe, I.M., Bassanelli, A., Puga, A., I. Papazoglou, I.,Zuchuat, O., Cazzoli , E., Eyink, J., Guida, G., PinolaL., Pulkkinen, U., Simola, K., von Winterfeldt, D.,Valeri, A. – Benchmark Exercise on expert judgementtechniques in PSA level 2 – To be published in theJournal Nuclear Engineering & Design.

14.40 Cojazzi, G., Fogli, D. (Eds.) with contributions from:STUK, VTT, ISTec, IPSN, Siemens. (2000), Bench-mark Exercise on Safety Evaluation of ComputerBased Systems (BE-SECBS Project), A shared-costaction Proposal under the 5th Framework Programmeof the European Atomic Energy Community(Euratom) for Research and Traning Activities, EC-JRC-ISIS, Technical Note No.I.00.96, Issue (3), July2000.

14.41 Kirchsteiger C., Cojazzi G. – Promotion of TechnicalHarmonisation on Risk-Based Decision Making –Proceedings of a Workshop held on May 22-24,2000, Grand Hotel Bristol, Stresa, Italy, 2 Vol.,EC-JRC S.P.I. 0063, May 2000.

14.42 Kirchsteiger C., Cojazzi G. – Promotion of TechnicalHarmonisation on Risk Based Decision Making –PSAM 5 International Conference on ProbabilisticSafety Assessment and Management, Osaka, Japan,27 November - 1 December 2000.

14.10 The European thematic network CERTA.

Società Italiana EsperienzeTermoidrauliche

ORGANISATION ACRONYM COUNTRY PROGRAMME

EC-JRC-ISIS Ispra, Italy LOBI

DIMNP-UoP Italy PIPERONE

GRS Germany ATHLET

Siemens AG, Power Generation (KWU) Siemens Germany PKL - UPTF

KFKI Atomic Energy Research Institute KFKI-AEKI Hungary PMK

Technical Research Centre of Finland VTT Finland PACTEL

SIET Italy SPES

Commissariat a l’Energie Atomique CEA France

Studsvik Eco & Safety AB EcoSafe Sweden FIX-II

Paul Scherrer Institute PSI Switzerland PANDA

Dipartimento Ingegneria Meccanica,Nucleare e Produzione

Università di Pisa

Gesellschaft für Anlagen- undReaktorsicherheit mbH

European Commission - J .R. C.,Institute for Systems,

Informatics and Safety

BETHSY-CATHARE

106

BWRs) as well as in the EU Enlargement countries (i.e.,VVERs). The CERTA network, which includes 10 majorEuropean research organisations, has been promotedand is being co-ordinated by ISIS (Figure 14.10).

14.5.2 ASTERISM2A.V. Jones, M. Delaval

The objective of this concerted action project is to im-prove the software developed by ISIS during the FourthFramework Programme (FP4) in order to produce aneffective and user-friendly database for the results ofresearch programmes. A trial version of the software isnow available on CD-ROM, charged with data from theN Research DG-funded nuclear fission safety projectsof FP4 concerned with source term phenomena. A userguide is provided as well as a (paper and on-screen)catalogue of the database contents, and the suppliersof the original data are now testing the database thor-oughly. For each project one may view the final reportand final summary report, tables of results, graphsand other graphics, and pieces of software where ap-propriate. The software is so documented that it canbe readily reused in other applications. The remainderof the project will examine extension to other nuclearsafety projects and links with existing databases of se-vere accident data, as well as make any necessary cor-rections to the database.

14.5.3 Enlargement

C. Addabbo, A. AnnunziatoTaking into account that the safety aspects connectedwith the reactors in operation in the Central and EasternEuropean Countries (CEEC) and in the New IndependentStates (NIS) are at the forefront of the EU enlargementagenda, exchange of information with CEEC and NISresearch organizations has been pursued. As part ofthe technical exchange agreement concluded with theElectrogorsk Research and Engineering Centre (EREC)of the Russian Federation, it has been envisaged tobenchmark the predictive capabilities of the EREC meltcoolant interaction code VAPEX against the test re-sults of FARO L-33 test results. L-33 has been the lasttest performed in the FARO test facility under extremeadverse conditions in order to trigger with an externalexplosive impulse (trigger) an energetic interaction.The results from test L-33 are rather unique calling fora deep and exhaustive analysis. This study will providean opportunity to share and transfer best practices inreactor severe accident safety analysis at the Europeanlevel through the appraisal of the predictive capabili-ties of the VAPEX code developed for VVER safety anal-ysis and their comparison with the results of the JRCCOMETA calculated results.

Within the long standing collaborative agreement be-tween JRC and the KFKI Atomic Energy Research Insti-

tute of Hungary, it has been agreed to carry out a studyto identify and apply multivariable signature analysismethodologies to discriminate accidents and transientsin PWRs and VVERs. This study will use a as test benchthe experimental results acquired in the JRC LOBI andin the KFKI PMK test facilities.

14.6 SPIN-OFF ACTIVITIES

14.6.1 Project Epsilon

B. WorthThis exploratory research builds upon theoretical dev-elopments in the area of two-phase flow and on a pre-vious shared-cost action project related to improvementsin the design of chemical reactors. Recognising that two-phase natural convection is a very important mechanismfor heat removal from a partially-voided nuclear reactorcore during accident conditions, Project Epsilon is look-ing into the application of “spin-enhanced naturalcirculation” also for non-nuclear process intensification.

The main objectives of Project Epsilon are to study theapplication of strong centrifugal force for enhancingelectrochemical reactions in a multiphase mixture,particularly for the generation of “clean energy” in anelectrochemical cell. Theoretical studies indicate thatmany multiphase processes which depend upon inter-facial mixing for mass and energy transfer, as well aschemical and/or electrochemical reactions at a cata-lytic boundary surface, are often diffusion-limitedprocesses. Gas generation or absorption at the surfaceof an electrode or within a porous bed in contact withliquid electrolyte, for example, depends on a combina-tion of fluid flow parameters generally controlled bythe interphase buoyancy term. In natural convectionprocesses, this depends crucially on the density differ-ence between two or more fluid phases (often a liquidand a gas), as well as on the local acceleration due togravity. For a given fluid mixture, the interphase buoy-ancy force can be increased substantially by the appli-cation of centrifugal forces, as occurs for example in acentrifuge. Spin-enhanced natural circulation can beused to greatly increase the thermodynamic efficiencyof many multiphase flow reaction processes, leadingto compact and hence cheaper designs of reactors, aswell as offering many advantages in terms of safetyover conventional fuel cell systems.

Project Epsilon is using the techniques of process in-tensification to design and construct an innovativeprototype centrifugal fuel cell for the clean generationof electricity. It has important links with nuclear ener-gy in respect to the fuel cycle of electropositive metalssuch as zinc and aluminium. With several patents pend-ing in Europe and elsewhere, it is hoped that this activitywill open up new avenues of EU research into innovativerenewable energy systems.


14.6.2 SUB-AERO

Y. DrossinosThe objective of this three-year, university-basedshared-cost action project is the investigation of subgrid(local) variability on ozone and fine particular matter inthe Mediterranean area. As a first step towards under-standing the effect of local variability on new particleformation, the effect of translational invariance onclassical homogeneous nucleation theory was investi-gated. This was done by proposing a connection bet-ween field-theoretic descriptions of condensation anddensity-functional theories of nucleation [14.43]. Theconnection was obtained starting from a quantum me-chanical Hamiltonian and using methods developed inthe context of studies of Bose-Einstein condensation[14.44].

14.6.3 LUNG

A.V. Jones, M. ReeksThe LUNG exploratory research project has the objec-tive of applying existing expertise in aerosol transportto improve understanding of particulate deposition inthe human respiratory system and hence of the hazardarising from airborne particulates. A comprehensivedeposition model which is frequently employed as abenchmark has been published by the InternationalCommittee for Radiation Protection, and a critical reviewhas been made of the modelling of aerosol depositionin the various components (especially the bronchialregion and alveolar interstitial regions) [14.45]. It wasconcluded that the modelling is relatively crude as re-gards the prediction of local surface concentrations ofdeposits e.g. at junctions, which are thought to play animportant role in the toxicology. A CFD/particle track-ing study has begun to examine deposition in such re-gions, using detailed calculations of the (normallylaminar) flow field to drive the particle motion. Impor-tant for fine particles is Brownian motion, which is sim-ulated by a Langevin term with a time scale on the orderof 1 % of the particle Stokes relaxation time.

The expertise acquired in LUNG has attracted interestfrom proposers of environmental and health-relatedshared-cost action projects. The contract for a projecton urban aerosol generation, transport and healthimpact is ready for signature, and negotiations are inprogress for a shared-cost action proposal on atmos-pheric particles and asthma.

14.6.4 Predicting Internet Traffic

Y. DrossinosThe objective of this exploratory research project wasto study and model the self-similar (non-Poissonian)structure of information flow in the Internet. Statisticalanalyses of Internet traffic have shown that conven-tional assumptions based on Poissonian statistics do

not reproduce the observed traffic flow. The associatedtime series (expressed, for example, as bytes per secondpassing through a server) have long-range correlationsand they appear unchanged under different levels ofaggregation: they are self-similar [14.46]. The resultsof a statistical analysis of access data of a JRC webserver (ARI FP4) are shown in Figure 14.11. The log-logfigure shows that times between successful URL re-quests (inactive periods) are distributed according toan algebraic distribution. The modelling work concen-trated on relating the power spectrum of a characteristictime series of packet traffic to the probability distribu-tions of active (ON) and inactive (OFF) periods. As shownin the figure, the inactive periods have a heavy tailed(Pareto-like) distribution; previous analyses of theactive periods resulted in similar distributions [14.46].The exponent characterising the low-frequency diver-gence of the power spectrum was related to propertiesof these distributions. Based on this relation an ana-lytic expression for the Hurst exponent, a quantity thatcharacterises persistence of correlations, was obtained[14.47].

References

14.43 Drossinos Y., Kevrekidis P.G., Lazaridis M.,Georgopoulos P.G. – Translational invariance in thetheory of nucleation – in “Nucleation and Atmos-pheric Aerosols 2000”, Editors B.N Hale and M.Kulmala, p. 217, AIP Conference Proceedings 2000.

14.44 Drossinos Y., Kevrekidis P.G., Georgopoulos P.G.– Translational invariance in theories of nucleation:Theoretical formulation – Physical Review E(in press) 2000.

14.45 Reeks, M.R. – Review of the ICRP Model –(in preparation).

14.46 Willinger W., Paxson V. – Where mathematicsmeets the Internet – Notices of the American Math-ematical Society, pp. 961-970, September 2000.

14.47 Haehner P., Drossinos Y. – A phenomenologicalapproach to intermittency – manuscript, 2000.

14.11 Statistical analysis of time periods between successful

URL requests for a JRC web server (ARI FP4).

Complementary distribution functionof time difference of (successful) URL requests

Complementary distribution functionof time difference of (successful) URL requests

time difference between (successful) requests in seconds

Com

plem

enta

ry d

istr

ibut

ion

func

tion

[Log

10(P

(X ›

x))] 100

10-1

10-2

10-3

10-4

10-5

10-6

10-7

10-8

100000100001000100101

13 sec ‹ t_{off} ‹ 15,000 sec

Some problems due toserver maintenance

Fit: 1-F (x)~x**(-alpha) with alpha = 0.83Web access data: 1-F(x)

108 Resources

Temporary Agents(maximum 3 years)

18

O÷cials131

Temporary Agents(5 years or indeterminate)

95

Auxiliaries46

Grantholders32

Visiting Scientifics7

Seconded National Experts

2

STAFF

BUDGET

ISIS obtains about 20 % of its funding from sources outside the direct actions of the Framework Programme (i.e.,institutional funding). These sources of complementary funding can be classified as:

• Competitive support to the Commission – was a mechanism employed during the Fourth Framework Programmewhereby a part of the JRC’s budget was put out to tender on projects selected by the Commission DGs. This endedwith the beginning of the Fifth Framework Programme so income from this source is zero for 2000.

• Other competitive activities – include projects paid by the DGs and Services of the European Commission directlyfrom their budget lines.

• Shared-cost actions – where the JRC pays 50 % of the costs from its own budget although some concerted actionsand supplementary measures are fully paid.

• Training and mobility of researchers – include individual fellowships from the Marie-Curie scheme and networksorganised around facilities such as ELSA.

• Third party work – includes work done for third parties.

Staff breakdown at the end of 2000.

* includes JRC matching funds.

Tables I, II and III below show details of income and work executed.

TYPE OF ACTIVITY FP5 (1999-2002) 2000

(M€) (M€)

Nuclear safeguards, non-proliferation

and international humanitarian security

Cyber-security, knowledge access and

assessment, the fight against fraud

Natural and technological risks and

emergencies

93.106 23.199

42.066 9.130

42.311 11.188

Table III: Institutional work(direct actions of the Framework Programme)

TYPE OF ACTIVITY 2000

(M€)

Competitive support to the E.C. 0.00

Other competitive activities 5.04

Shared-cost actions* 4.60

Training and mobility of researchers 0.18

Third-party work 1.50

TOTAL 11.32

Table I: Contracts signed.

TYPE OF ACTIVITY 2000

(M€)

Competitive support to the E.C. 3.93

Other competitive activities 2.75

Shared-cost actions* 3.95

Training and mobility of researchers 0.20

Third-party work 1.58

TOTAL 12.41

Table II: Competitive work.

RESOURCES

109

ISIS CONTACT POINTS (corresponding to first author)

E-mail: [email protected]

http://www.jrc.cec.eu.int/isis/

Tel: +39 0332 78 9322/9511

Addabbo, Carmelo

Nuclear [email protected]: +39 0332 789812

Al-Khudhairy, Delilah

Methodologies for Information [email protected]: +39 0332 785696

Andritsos, Fivos

Reliable Information [email protected]: +39 0332 789599

Annunziato, Alessandro


Arsenis, Spyros

Risk Management and Decision [email protected]: +39 0332 785357

Bain, Donald


Barbas, Thomas


Berndt, Reinhard

Safeguards and Verification [email protected]: +39 0332 785317

Best, Clive


Breitenbach, Lothar


Cacciabue, Pietro Carlo


Campolongo, Francesca


Casadei, Folco

Safety in Structural [email protected]: +39 0332 789563

Cojazzi, Giacomo


Colombo, Alessandro


Cuypers, Marc (Unit Head)Safeguards and Verification [email protected]: +39 0332 789515

d’Agraives, Bertrand

Safeguards and Verification TechniquesBertrand.d’[email protected]: +39 0332 789107

Dowell, Martyn (Unit Head)Methodologies for Information [email protected]: +39 0332 789251

Drossinos, Ioannis


Duffield, Stuart


Franklin, Michael


Funtowicz, Silvio


Géradin, Michel (Unit Head)Safety in Structural [email protected]: +39 0332 789989

Gonçalves, João


Grossetie, Jean-Claude


Gutiérrez, Eugenio


Huld, Thomas


Hunt, Brian


Jackson, Tom


Jones, Alan


Korn, Christophe


Littman, Francois


Loekkemyhr, Per


Lucia, Alfredo (Unit Head)Risk Management and Decision [email protected]: +39 0332 789155

Magonette, Georges


Masera, Marcelo


ISIS Contact Points

110

Mazzeo Rinaldi, Francesco


Millán Ruiz, José del R.


Negro, Paolo


Nordvik, Jean-Pierre


Peckham, Robert


Pegon, Pierre


Perdigão, José


Pinto, Artur

Safety in Stuctural [email protected]: +39 0332 789294

Pirelli, Giuliano


Planas, Christophe


Post, Wietse


Poucet, André


Reggiori, Alberto


Renda, Vito


Rossi, Alessandro


Ruiz Morales, Emilio


Saltelli, Andrea


Schillebeeckx, Peter


Shepherd, Iain


Sequeira, Vitor


Sieber, Alois (Unit Head)Technologies for Detection & [email protected]: +39 0332 789089

Solomos, George


Sorel, Fernand (Unit Head)Reliable Information [email protected]: +39 0332 789411

Städtke, Herbert


Steinberger, Ralf


Stringa, Elena


Tarchi, Dario

Technologies for Detection & [email protected]: +39 0332 785143

Tarantola, Stefano


Weisshäupl, Horst (Unit Head)Nuclear [email protected]: +39 0332 785585

Wider, Hartmut


Wilkens, Marc


Wilkinson, David

Director of ISIS

[email protected]: +39 0332 789947

Worth, Brian


Zeyen, Roland


ISIS Contact Points

European Commission

EUR Report 19743 EN – Institute for Systems, Informatics and Safety (ISIS) – Annual Report 2000

Editor: Anne-Marie Morrissey

Luxembourg: Office for Official Publications of the European Communities

2001 – 112 pp. – 21.0 x 29.7 cm

Scientific and Technical Research series

ISBN 92-894-0602-X

ACKNOWLEDGEMENTS

The editor would like to take this opportunity to thank all the staff who contributed to the Annual Report and to theUnit Heads for co-ordinating the response of their respective units. A special thanks is extended to Ramón Sánchezfor the layout and presentation.

111

15LB

-NA

-19-7

43

-EN

-C

The mission of the JRC is to provide customer-driven scientific and technical support for the conception, develop-ment, implementation and monitoring of EU policies. As a service of the European Commission, the JRC functionsas a reference centre of science and technology for the Union. Close to the policy-making process, it serves thecommon interest of the Member States, while being independent of special interests, whether private or national.

EUROPEAN COMMISSIONJOINT RESEARCH CENTRE